Skip to page content
SAVRN
Search Plan a project

Open-weight model

routing-analysis-anvil-six-models

by Liang Liang lylybig8/routing-analysis-anvil-six-models

routing-analysis-anvil-six-models is an open-weight model from Liang Liang. Its published files total 1.1 TB.

This repository is a path-preserving research snapshot for Hausa and Yoruba adaptation experiments based on AIDC-AI/Marco-Nano-Base.

Parameters—
Context—
Weights1.1 TB
License—
AccessOpen weights
Monthly Downloads—

Model Card

This repository is a path-preserving research snapshot for Hausa and Yoruba adaptation experiments based on AIDC-AI/Marco-Nano-Base. It contains model weights, resumable checkpoints, training and evaluation code, logs, and raw LightEval results collected on the Anvil cluster. The directory hierarchy follows the original routinganalysis project so that scripts and recorded paths remain easy to trace. No symbolic links are - finetuning/phase2full/checkpoints/marconanobase/hauLatn100kfull - finetuning/phase2full/checkpoints/marconanobase/yorLatn50kfull - finetuning/phase2full/checkpoints/marconanobase/yorLatn50kexp3trainonlyyorpretrainingcorpustop8experts These directories are preserved in…

Excerpt from the card by Liang Liang.

Identity and Version

Repository
lylybig8/routing-analysis-anvil-six-models
Publisher
Liang Liang
Task
Not stated by the source
Modality
Other
Library
transformers
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
99ceb0394fc30382a56fb1e1bd4672b18c037059
First published
2026-10-03
Last updated
2026-10-03

Files and Weights

1,163 files, 1.1 TB in total. The weights are 204 files totalling 1.1 TB in bin, pt, pth, safetensors.

Weights204 files · 1.1 TB
Configuration559 files · 65.0 MB
Tokenizer59 files · 227.7 MB
Documentation14 files · 153.3 KB
Other320 files · 37.0 GB
Repository7 files · 20.5 MB
Every file
FileTypeSizeSHA-256
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/adapter_model.safetensorsWeights242.4 MB aea9ceea45b2
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/optimizer.ptWeights485.6 MB a9973ec63660
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/rng_state_0.pthWeights15.4 KB 9a62337cc95e
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/rng_state_1.pthWeights15.4 KB 63a38473375a
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/rng_state_2.pthWeights15.4 KB 104e35fe44a9
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/rng_state_3.pthWeights15.4 KB e18e85c5c0a4
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/scheduler.ptWeights1.5 KB cd574313c6d2
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/training_args.binWeights6.2 KB e35409a52440
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00001-of-00007.safetensorsWeights5.0 GB c842902195a8
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00002-of-00007.safetensorsWeights5.0 GB 2523eee92590
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00003-of-00007.safetensorsWeights5.0 GB 4a492c11a7e0
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00004-of-00007.safetensorsWeights5.0 GB 650986cd8500
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00005-of-00007.safetensorsWeights5.0 GB 0dcd0c5ea130
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00006-of-00007.safetensorsWeights5.0 GB dd1eb10f811f
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model-00007-of-00007.safetensorsWeights2.6 GB 7b1d38392b0a
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/optimizer.binWeights64.0 GB 34a9ac0fbf3d
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/pytorch_model_fsdp.binWeights32.6 GB b51bfb46717c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/scheduler.ptWeights1.5 KB a9a5bc58d777
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/training_args.binWeights6.2 KB ae1bde3f0e28
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00001-of-00007.safetensorsWeights5.0 GB cdbee3e22ca3
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00002-of-00007.safetensorsWeights5.0 GB 2e8486cab6e9
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00003-of-00007.safetensorsWeights5.0 GB 8adf8379a9ea
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00004-of-00007.safetensorsWeights5.0 GB d3d5df562470
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00005-of-00007.safetensorsWeights5.0 GB 8762a21cb675
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00006-of-00007.safetensorsWeights5.0 GB e6a40ee7ff96
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model-00007-of-00007.safetensorsWeights2.6 GB 3d134067b7bc
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/optimizer.binWeights64.0 GB 759e309f1e96
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/pytorch_model_fsdp.binWeights32.6 GB c515ab0ae68c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/scheduler.ptWeights1.5 KB 07ae1dcc4e6d
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/training_args.binWeights6.2 KB ae1bde3f0e28
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00001-of-00007.safetensorsWeights5.0 GB cdbee3e22ca3
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00002-of-00007.safetensorsWeights5.0 GB 2e8486cab6e9
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00003-of-00007.safetensorsWeights5.0 GB 8adf8379a9ea
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00004-of-00007.safetensorsWeights5.0 GB d3d5df562470
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00005-of-00007.safetensorsWeights5.0 GB 8762a21cb675
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00006-of-00007.safetensorsWeights5.0 GB e6a40ee7ff96
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model-00007-of-00007.safetensorsWeights2.6 GB 3d134067b7bc
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/training_args.binWeights6.2 KB ae1bde3f0e28
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00001-of-00007.safetensorsWeights5.0 GB 8e35a5b5fcee
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00002-of-00007.safetensorsWeights5.0 GB dd824c9840ab
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00003-of-00007.safetensorsWeights5.0 GB ef272b99aa1f
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00004-of-00007.safetensorsWeights5.0 GB fbcb5a43af3d
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00005-of-00007.safetensorsWeights5.0 GB 14134c45ebce
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00006-of-00007.safetensorsWeights5.0 GB 0f86a15b9724
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model-00007-of-00007.safetensorsWeights2.0 GB 5a4228b1ba4c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/optimizer.binWeights2.1 GB b92140a57ab9
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/pytorch_model_fsdp.binWeights32.0 GB 4507b7c85d04
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_4.pthWeights15.4 KB 68e2197b24fb
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_5.pthWeights15.4 KB aa9deb9b7c31
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_6.pthWeights15.4 KB 12bbe3cda4e3
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/rng_state_7.pthWeights15.4 KB 234cf1c0e059
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/scheduler.ptWeights1.5 KB f8a9a1ef3783
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/training_args.binWeights6.4 KB 6eb4ff018bbb
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/optimizer_rank_00000_of_00004.ptWeights16.0 GB 25d0a4a6d8a1
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/optimizer_rank_00001_of_00004.ptWeights16.0 GB d091605135b9
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/optimizer_rank_00002_of_00004.ptWeights16.0 GB 9c260b839ff3
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/optimizer_rank_00003_of_00004.ptWeights16.0 GB 1655b6e47ce2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/scheduler.ptWeights1.5 KB 93ca4aa34b9a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00001-of-00007.safetensorsWeights5.0 GB 7c35abaedf50
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00002-of-00007.safetensorsWeights5.0 GB d3226bfd35be
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00003-of-00007.safetensorsWeights5.0 GB 6a518961c048
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00004-of-00007.safetensorsWeights5.0 GB 27b20a1b10bc
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00005-of-00007.safetensorsWeights5.0 GB 5b5d460944cc
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00006-of-00007.safetensorsWeights5.0 GB 356144678577
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model-00007-of-00007.safetensorsWeights2.0 GB 8caccbe8924d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/optimizer.binWeights2.1 GB 51c0a8b4ae7e
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/pytorch_model_fsdp.binWeights32.0 GB 434a2bc289e8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/scheduler.ptWeights1.5 KB e43e4d046d52
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00001-of-00007.safetensorsWeights5.0 GB 8b6cad7f443f
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00002-of-00007.safetensorsWeights5.0 GB 7271c27c5d57
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00003-of-00007.safetensorsWeights5.0 GB ab3898be8fff
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00004-of-00007.safetensorsWeights5.0 GB 4f13bce6008d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00005-of-00007.safetensorsWeights5.0 GB 6a83cd2c8573
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00006-of-00007.safetensorsWeights5.0 GB 940f4aefe474
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model-00007-of-00007.safetensorsWeights2.0 GB d97451adfcb2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/optimizer.binWeights2.1 GB d2f28320ee86
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/pytorch_model_fsdp.binWeights32.0 GB a022e2a409f3
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/scheduler.ptWeights1.5 KB b65707c1cd19
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00001-of-00007.safetensorsWeights5.0 GB 829e85c3dfde
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00002-of-00007.safetensorsWeights5.0 GB cead5d7b223f
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00003-of-00007.safetensorsWeights5.0 GB 3d9c2257e908
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00004-of-00007.safetensorsWeights5.0 GB 9e055c85606e
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00005-of-00007.safetensorsWeights5.0 GB 205390603973
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00006-of-00007.safetensorsWeights5.0 GB 3aff93fc567f
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model-00007-of-00007.safetensorsWeights2.0 GB 62885cd87b36
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/optimizer.binWeights2.1 GB eb02da4fbc45
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/pytorch_model_fsdp.binWeights32.0 GB c061a64bf438
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/scheduler.ptWeights1.5 KB fe32e31c2b0e
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00001-of-00007.safetensorsWeights5.0 GB 7ca239df27ae
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00002-of-00007.safetensorsWeights5.0 GB 7fc8e21185bb
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00003-of-00007.safetensorsWeights5.0 GB 1aff4f83ef3e
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00004-of-00007.safetensorsWeights5.0 GB a8c8aa6ac4ad
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00005-of-00007.safetensorsWeights5.0 GB a296838bf605
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00006-of-00007.safetensorsWeights5.0 GB 90b6fe8f5b93
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model-00007-of-00007.safetensorsWeights2.0 GB 280086fd72a1
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/optimizer.binWeights2.1 GB 532125e0a005
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/pytorch_model_fsdp.binWeights32.0 GB c7f1908c1e31
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/scheduler.ptWeights1.5 KB c96b4ddc1727
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00001-of-00007.safetensorsWeights5.0 GB a13435fb7c35
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00002-of-00007.safetensorsWeights5.0 GB b9b316e39b2d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00003-of-00007.safetensorsWeights5.0 GB c4ecdc2010b7
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00004-of-00007.safetensorsWeights5.0 GB 6a2b8dc37285
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00005-of-00007.safetensorsWeights5.0 GB 86c5e742da2a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00006-of-00007.safetensorsWeights5.0 GB 570eb02631b2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model-00007-of-00007.safetensorsWeights2.0 GB 9330559bb20a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/optimizer.binWeights2.1 GB ded7bf24d7d1
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/pytorch_model_fsdp.binWeights32.0 GB cf084f1bb54d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/scheduler.ptWeights1.5 KB 4e1a1981fc2a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00001-of-00007.safetensorsWeights5.0 GB a13435fb7c35
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00002-of-00007.safetensorsWeights5.0 GB b9b316e39b2d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00003-of-00007.safetensorsWeights5.0 GB c4ecdc2010b7
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00004-of-00007.safetensorsWeights5.0 GB 6a2b8dc37285
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00005-of-00007.safetensorsWeights5.0 GB 86c5e742da2a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00006-of-00007.safetensorsWeights5.0 GB 570eb02631b2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model-00007-of-00007.safetensorsWeights2.0 GB 9330559bb20a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/training_args.binWeights6.4 KB b1094eb31bb8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00001-of-00007.safetensorsWeights5.0 GB 03bebc8b5b34
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00002-of-00007.safetensorsWeights5.0 GB 5dacc1606ff4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00003-of-00007.safetensorsWeights5.0 GB 363cfde7a7ae
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00004-of-00007.safetensorsWeights5.0 GB 210ede9cbce9
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00005-of-00007.safetensorsWeights5.0 GB f701e66309f6
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00006-of-00007.safetensorsWeights5.0 GB 28525b5b1c9b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model-00007-of-00007.safetensorsWeights2.6 GB f95228d4b48d
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/optimizer.binWeights64.0 GB d7a728ba0b16
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/pytorch_model_fsdp.binWeights32.6 GB 437c1bc580f9
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/scheduler.ptWeights1.5 KB b65707c1cd19
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/training_args.binWeights6.2 KB eb9c3c0cbd01
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00001-of-00007.safetensorsWeights5.0 GB b14b0177c3b2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00002-of-00007.safetensorsWeights5.0 GB 53ee42366569
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00003-of-00007.safetensorsWeights5.0 GB 819aa0cd1283
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00004-of-00007.safetensorsWeights5.0 GB 57e6cef038c1
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00005-of-00007.safetensorsWeights5.0 GB 856a86442fea
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00006-of-00007.safetensorsWeights5.0 GB eec8dd2beddb
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model-00007-of-00007.safetensorsWeights2.6 GB 81a2cfeb4a91
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/optimizer.binWeights64.0 GB d6bd8318c2d2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/pytorch_model_fsdp.binWeights32.6 GB 37870040c2a8
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/rng_state_0.pthWeights15.4 KB 659b1cdee221
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/rng_state_1.pthWeights15.4 KB 86accf27064c
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/rng_state_2.pthWeights15.4 KB 18ca8d714ef4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/rng_state_3.pthWeights15.4 KB 2cfdebe99e40
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/scheduler.ptWeights1.5 KB 4e1a1981fc2a
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/training_args.binWeights6.2 KB eb9c3c0cbd01
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00001-of-00007.safetensorsWeights5.0 GB b14b0177c3b2
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00002-of-00007.safetensorsWeights5.0 GB 53ee42366569
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00003-of-00007.safetensorsWeights5.0 GB 819aa0cd1283
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00004-of-00007.safetensorsWeights5.0 GB 57e6cef038c1
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00005-of-00007.safetensorsWeights5.0 GB 856a86442fea
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00006-of-00007.safetensorsWeights5.0 GB eec8dd2beddb
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model-00007-of-00007.safetensorsWeights2.6 GB 81a2cfeb4a91
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/training_args.binWeights6.2 KB eb9c3c0cbd01
models/Marco-Nano-Base/model-00001-of-00009.safetensorsWeights2.0 GB 43928a210920
models/Marco-Nano-Base/model-00002-of-00009.safetensorsWeights2.0 GB c9a585d1032a
models/Marco-Nano-Base/model-00003-of-00009.safetensorsWeights2.0 GB 2e09c8f17f29
models/Marco-Nano-Base/model-00004-of-00009.safetensorsWeights2.0 GB ae23ebbc85dc
models/Marco-Nano-Base/model-00005-of-00009.safetensorsWeights2.0 GB 639923c4f1a8
models/Marco-Nano-Base/model-00006-of-00009.safetensorsWeights2.0 GB 8eccab512ec9
models/Marco-Nano-Base/model-00007-of-00009.safetensorsWeights2.0 GB 64db2f612bde
models/Marco-Nano-Base/model-00008-of-00009.safetensorsWeights2.0 GB 083855cc8286
models/Marco-Nano-Base/model-00009-of-00009.safetensorsWeights318.3 MB 790390ae35ce
Marco-LLM-eval-main/base_tasks.pyConfiguration72.3 KB —
finetuning/data/splits/hau_Latn_manifest.jsonConfiguration2.0 KB —
finetuning/data/splits/yor_Latn_manifest.jsonConfiguration1.3 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/pre_domain_review_v1/yor_Latn_100k_fw2_wura_manifest.jsonConfiguration3.8 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/pre_domain_review_v1/yor_Latn_100k_fw2_wura_selection.jsonConfiguration14.3 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/pre_domain_review_v1/yor_Latn_100k_fw2_wura_verification.jsonConfiguration14.5 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/yor_Latn_100k_fw2_wura_manifest.jsonConfiguration4.0 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/yor_Latn_100k_fw2_wura_selection.jsonConfiguration15.3 KB —
finetuning/data/yoruba_100k_fw2_wura/outputs/yor_Latn_100k_fw2_wura_verification.jsonConfiguration13.9 KB —
finetuning/phase1_lora/finetune_lora.pyConfiguration28.7 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/TIME_LIMIT_STOP_20260930T023053Z.jsonConfiguration431 B —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/adapter_config.jsonConfiguration33.0 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/added_tokens.jsonConfiguration707 B —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/special_tokens_map.jsonConfiguration613 B —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/trainer_state.jsonConfiguration54.8 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/run_config.jsonConfiguration3.0 KB —
finetuning/phase1_lora/top8_expert_lora/finetune_lora_top8_experts.pyConfiguration23.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/trainer_state.jsonConfiguration243.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/trainer_state.jsonConfiguration253.3 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/run_config.jsonConfiguration790 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/HAU_500K_RAW_CACHE_READY_20260930T215925Z-pid785986.jsonConfiguration79 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/HAU_500K_RESUME_SELECTION_20260930T215925Z-pid785986.jsonConfiguration223 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/HAU_500K_RUN_BINDING.jsonConfiguration3.7 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/HAU_500K_TOKENIZED_CACHE_READY_20260930T215925Z-pid785986.jsonConfiguration294 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/TRAINING_PAUSED.jsonConfiguration452 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank0.jsonConfiguration41.1 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank1.jsonConfiguration41.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank2.jsonConfiguration42.1 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank3.jsonConfiguration42.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank4.jsonConfiguration41.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank5.jsonConfiguration43.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank6.jsonConfiguration42.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/audit_rank7.jsonConfiguration43.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/HAU_500K_CHECKPOINT_COMPLETE.jsonConfiguration6.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/trainer_state.jsonConfiguration274.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/freeze_summary.jsonConfiguration42.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/run_config.jsonConfiguration45.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/trainer_state.jsonConfiguration274.7 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/FSDP_SHARDED_CHECKPOINT_COMPLETE.jsonConfiguration259 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/trainer_state.jsonConfiguration84.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/run_config.jsonConfiguration821 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/TRAINING_COMPLETE.jsonConfiguration3.1 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/TRAIN_ADAPTER_COMPLETE.jsonConfiguration260 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/YOR_RESUME_SELECTION_20260926T210504Z-production_782.jsonConfiguration63 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/YOR_RUN_BINDING.jsonConfiguration2.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/audit_rank0.jsonConfiguration44.9 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/audit_rank1.jsonConfiguration46.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/audit_rank2.jsonConfiguration48.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/audit_rank3.jsonConfiguration47.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/YOR_CHECKPOINT_COMPLETE.jsonConfiguration4.8 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/trainer_state.jsonConfiguration65.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/YOR_CHECKPOINT_COMPLETE.jsonConfiguration4.8 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/trainer_state.jsonConfiguration81.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/YOR_CHECKPOINT_COMPLETE.jsonConfiguration4.8 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/trainer_state.jsonConfiguration97.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/YOR_CHECKPOINT_COMPLETE.jsonConfiguration4.8 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/trainer_state.jsonConfiguration113.7 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/YOR_CHECKPOINT_COMPLETE.jsonConfiguration4.8 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/trainer_state.jsonConfiguration127.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/freeze_summary.jsonConfiguration42.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/run_config.jsonConfiguration45.2 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/trainer_state.jsonConfiguration127.3 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/trainer_state.jsonConfiguration80.9 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/added_tokens.jsonConfiguration707 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/trainer_state.jsonConfiguration126.5 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/config.jsonConfiguration1.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/generation_config.jsonConfiguration121 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/model.safetensors.index.jsonConfiguration1.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/run_config.jsonConfiguration789 B —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/special_tokens_map.jsonConfiguration613 B —
finetuning/phase2_full/finetune_full.pyConfiguration12.5 KB —
finetuning/phase2_full/finetune_full_freeze.pyConfiguration19.1 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/scripts/check_2node_nccl.pyConfiguration1.7 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/train_hau_500k_top8.pyConfiguration60.3 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/common.pyConfiguration7.8 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/data/mode2_outputs/activation_counts_pretraining_corpus_marco_nano_base_hau_hau_Latn_seed42_1lang.jsonConfiguration10.1 MB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/generate_freeze_plan.pyConfiguration11.7 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/plans/PLAN_VALIDATED.jsonConfiguration522 B —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/plans/exp3_train_only_top8_hau_pretraining_corpus_50k_experts.jsonConfiguration60.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/validate_checkpoints.pyConfiguration14.1 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/common.pyConfiguration8.1 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/.prepare_binding.jsonConfiguration4.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/PREPARATION_COMPLETE.jsonConfiguration1.6 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/belebele_yor_Latn_mode2_length_targets.jsonConfiguration235.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/fineweb_yor_Latn_parquet_index.jsonConfiguration1.6 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/matched_yor_Latn_seed42.meta.jsonConfiguration4.5 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/mode2_outputs/activation_counts_pretraining_corpus_marco_nano_base_yor_yor_Latn_seed42_1lang.jsonConfiguration10.3 MB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/mode2_outputs/activation_counts_pretraining_corpus_marco_nano_base_yor_yor_Latn_seed42_1lang.meta.jsonConfiguration5.1 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/training_exclusion_manifest.jsonConfiguration14.2 MB 890aef994468
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/finalize_training.pyConfiguration14.2 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/generate_freeze_plan.pyConfiguration11.7 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/plans/PLAN_VALIDATED.jsonConfiguration522 B —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/plans/exp3_train_only_top8_yor_pretraining_corpus_50k_experts.jsonConfiguration60.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/prepare_calibration.pyConfiguration26.3 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/quarantine_incomplete_root.pyConfiguration3.0 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/run_mode2_routing.pyConfiguration13.9 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/train_yor_top8.pyConfiguration26.5 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/validate_checkpoints.pyConfiguration14.5 KB —
finetuning/phase2_full/hausa_100k_full/finetune_full_time_limited.pyConfiguration7.2 KB —
finetuning/phase2_full/hausa_100k_full/preflight_hausa_100k_full.pyConfiguration70.3 KB —
finetuning/phase2_full/hausa_100k_full/prepare_smoke_split.pyConfiguration2.1 KB —
finetuning/phase2_full/hausa_100k_full/verify_hausa_100k_full.pyConfiguration28.3 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.metadata.jsonConfiguration8.2 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260926T193710Z-2311252.event.10.torchrun_observed.20260926T193710Z-2311252.jsonConfiguration1.2 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T044737Z-2899040.event.10.torchrun_observed.20260928T044737Z-2899040.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T044737Z-2899040.event.20.preverification_failed.20260928T044737Z-2899040.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T045031Z-2900139.event.10.torchrun_observed.20260928T045031Z-2900139.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T045031Z-2900139.event.20.preverification_succeeded.20260928T045031Z-2900139.jsonConfiguration2.6 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T045031Z-2900139.event.30.strict_verification_succeeded.20260928T045031Z-2900139.jsonConfiguration2.6 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T045031Z-2900139.proof.20_preverification.jsonConfiguration4.6 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.20260928T045031Z-2900139.proof.30_strict_verification.jsonConfiguration4.9 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.status.jsonConfiguration1.0 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.time_limit_stop.20260926T193710Z-2311252.jsonConfiguration406 B —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.verification.jsonConfiguration4.9 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.metadata.jsonConfiguration7.5 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.metadata.pre-sharded-checkpoint-fix-20260930T210115EDT.jsonConfiguration8.6 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20260929T174115Z-688620.event.10.torchrun_observed.20260929T174115Z-688620.jsonConfiguration1.2 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20260930T183105Z-4007815.event.10.torchrun_observed.20260930T183105Z-4007815.jsonConfiguration2.2 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20260930T183105Z-4007815.event.20.preverification_failed.20260930T183105Z-4007815.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20261001T015238Z-4140225.event.10.torchrun_observed.20261001T015238Z-4140225.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20261001T015238Z-4140225.event.20.preverification_failed.20261001T015238Z-4140225.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.20261001T213326Z-306451.event.10.torchrun_observed.20261001T213326Z-306451.jsonConfiguration2.3 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.status.jsonConfiguration974 B —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.time_limit_stop.20260929T174115Z-688620.jsonConfiguration406 B —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.time_limit_stop.20261001T211336Z-dcp-gloo-validation.jsonConfiguration406 B —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.time_limit_stop.20261001T213326Z-306451.jsonConfiguration406 B —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.metadata.jsonConfiguration8.2 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.20260926T201522Z-3640026.event.10.torchrun_observed.20260926T201522Z-3640026.jsonConfiguration1.2 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.20260926T201522Z-3640026.event.20.preverification_succeeded.20260926T201522Z-3640026.jsonConfiguration2.5 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.20260926T201522Z-3640026.event.30.strict_verification_succeeded.20260926T201522Z-3640026.jsonConfiguration2.6 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.20260926T201522Z-3640026.proof.20_preverification.jsonConfiguration4.5 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.20260926T201522Z-3640026.proof.30_strict_verification.jsonConfiguration4.9 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.status.jsonConfiguration1.0 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.verification.jsonConfiguration4.9 KB —
finetuning/phase2_full/yoruba_100k_full/finetune_full_time_limited.pyConfiguration22.7 KB —
finetuning/phase2_full/yoruba_100k_full/preflight_yoruba_100k_full.pyConfiguration77.1 KB —
finetuning/phase2_full/yoruba_100k_full/prepare_smoke_split.pyConfiguration2.1 KB —
finetuning/phase2_full/yoruba_100k_full/test_dcp_gloo.pyConfiguration1.1 KB —
finetuning/phase2_full/yoruba_100k_full/verify_yoruba_100k_full.pyConfiguration28.3 KB —
finetuning/phase2_full/yoruba_50k_full/finetune_full_time_limited.pyConfiguration7.2 KB —
finetuning/phase2_full/yoruba_50k_full/preflight_yoruba_50k_full.pyConfiguration70.3 KB —
finetuning/phase2_full/yoruba_50k_full/prepare_smoke_split.pyConfiguration2.1 KB —
finetuning/phase2_full/yoruba_50k_full/verify_yoruba_50k_full.pyConfiguration28.3 KB —
models/Marco-Nano-Base/config.jsonConfiguration1.0 KB —
models/Marco-Nano-Base/configuration.jsonConfiguration48 B —
models/Marco-Nano-Base/generation_config.jsonConfiguration121 B —
models/Marco-Nano-Base/model.safetensors.index.jsonConfiguration1.8 MB —
reproduce/generate_global_mmlu_full_tables.pyConfiguration15.8 KB —
reproduce/global_mmlu_accelerated_3node/lib/stage_model_to_shm.pyConfiguration21.4 KB —
reproduce/global_mmlu_accelerated_3node/lib/validate_bundle.pyConfiguration24.4 KB —
reproduce/global_mmlu_accelerated_3node/lib/verify_completion.pyConfiguration30.7 KB —
reproduce/global_mmlu_tasks_plus_hau.pyConfiguration1.4 KB —
reproduce/lighteval_belebele_tasks.pyConfiguration2.4 KB —
reproduce/lighteval_belebele_tasks_hau.pyConfiguration2.4 KB —
reproduce/lighteval_belebele_tasks_yor.pyConfiguration2.4 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/arb_Arab/results_2026-09-29T14-08-15.217809.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/azj_Latn/results_2026-09-29T14-21-51.833801.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ben_Beng/results_2026-09-29T14-38-31.600954.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/cat_Latn/results_2026-09-29T14-51-48.165321.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ces_Latn/results_2026-09-29T15-07-30.842539.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/dan_Latn/results_2026-09-29T15-23-04.035220.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/deu_Latn/results_2026-09-29T15-38-06.398002.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ell_Grek/results_2026-09-29T15-55-38.667522.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/eng_Latn/results_2026-09-29T16-05-31.157833.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/fra_Latn/results_2026-09-29T14-08-15.217791.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/hau_Latn/results_2026-09-29T15-37-59.266006.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/heb_Hebr/results_2026-09-29T14-21-24.418755.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/hun_Latn/results_2026-09-29T14-36-38.029472.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ind_Latn/results_2026-09-29T14-51-50.977818.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ita_Latn/results_2026-09-29T15-07-16.674462.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/jpn_Jpan/results_2026-09-29T15-22-45.301888.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/kaz_Cyrl/results_2026-09-29T15-39-20.946918.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/kor_Hang/results_2026-09-29T15-53-38.410965.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/nld_Latn/results_2026-09-29T16-06-11.870528.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/npi_Deva/results_2026-09-29T14-10-08.633687.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/pes_Arab/results_2026-09-29T14-21-45.101266.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/pol_Latn/results_2026-09-29T14-36-32.400003.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/por_Latn/results_2026-09-29T14-51-44.748480.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ron_Latn/results_2026-09-29T15-07-25.070910.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/rus_Cyrl/results_2026-09-29T15-23-16.769077.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/spa_Latn/results_2026-09-29T15-38-03.985408.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/swe_Latn/results_2026-09-29T15-53-36.971288.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/tam_Taml/results_2026-09-29T14-17-25.805683.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/tha_Thai/results_2026-09-29T14-25-07.424756.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/tur_Latn/results_2026-09-29T14-36-32.731885.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/ukr_Cyrl/results_2026-09-29T14-52-11.242574.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/urd_Arab/results_2026-09-29T15-08-27.240910.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/vie_Latn/results_2026-09-29T15-22-42.649861.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/zho_Hans/results_2026-09-29T15-53-09.916511.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/zsm_Latn/results_2026-09-29T16-06-07.075230.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/arb_Arab/results_2026-09-28T03-59-20.757312.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/azj_Latn/results_2026-09-28T04-12-21.154208.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ben_Beng/results_2026-09-28T04-27-53.032955.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/cat_Latn/results_2026-09-28T04-40-23.796731.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ces_Latn/results_2026-09-28T04-54-59.046141.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/dan_Latn/results_2026-09-28T05-10-06.607564.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/deu_Latn/results_2026-09-28T05-24-08.322869.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ell_Grek/results_2026-09-28T05-40-33.804526.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/eng_Latn/results_2026-09-28T05-49-34.309246.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/fra_Latn/results_2026-09-28T03-59-21.777983.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/heb_Hebr/results_2026-09-28T04-11-49.077812.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/hun_Latn/results_2026-09-28T04-26-09.818379.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ind_Latn/results_2026-09-28T04-40-16.829756.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ita_Latn/results_2026-09-28T04-54-41.866811.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/jpn_Jpan/results_2026-09-28T05-09-50.738400.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/kaz_Cyrl/results_2026-09-28T05-25-27.894821.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/kor_Hang/results_2026-09-28T05-38-31.803848.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/nld_Latn/results_2026-09-28T05-50-15.760527.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/npi_Deva/results_2026-09-28T04-01-02.317309.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/pes_Arab/results_2026-09-28T04-12-18.111262.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/pol_Latn/results_2026-09-28T04-26-00.140685.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/por_Latn/results_2026-09-28T04-40-23.663001.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ron_Latn/results_2026-09-28T04-54-47.452367.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/rus_Cyrl/results_2026-09-28T05-10-05.210381.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/spa_Latn/results_2026-09-28T05-24-05.593806.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/swe_Latn/results_2026-09-28T05-38-21.997828.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/tam_Taml/results_2026-09-28T04-08-15.580693.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/tha_Thai/results_2026-09-28T04-15-35.553062.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/tur_Latn/results_2026-09-28T04-26-02.656575.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/ukr_Cyrl/results_2026-09-28T04-40-44.834234.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/urd_Arab/results_2026-09-28T04-55-52.791009.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/vie_Latn/results_2026-09-28T05-09-40.684650.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/yor_Latn/results_2026-09-28T05-24-30.777258.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/zho_Hans/results_2026-09-28T05-37-53.658498.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/zsm_Latn/results_2026-09-28T05-50-06.282033.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/arb_Arab/results_2026-09-28T04-07-21.321360.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/azj_Latn/results_2026-09-28T04-20-16.994575.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ben_Beng/results_2026-09-28T04-36-13.844010.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/cat_Latn/results_2026-09-28T04-48-32.944117.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ces_Latn/results_2026-09-28T05-03-21.050009.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/dan_Latn/results_2026-09-28T05-18-34.763716.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/deu_Latn/results_2026-09-28T05-32-55.682180.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ell_Grek/results_2026-09-28T05-49-16.437678.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/eng_Latn/results_2026-09-28T05-58-19.570927.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/fra_Latn/results_2026-09-28T04-07-24.080239.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/heb_Hebr/results_2026-09-28T04-19-54.022459.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/hun_Latn/results_2026-09-28T04-34-21.750979.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ind_Latn/results_2026-09-28T04-48-38.701308.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ita_Latn/results_2026-09-28T05-03-12.214179.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/jpn_Jpan/results_2026-09-28T05-18-20.054560.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/kaz_Cyrl/results_2026-09-28T05-34-03.971381.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/kor_Hang/results_2026-09-28T05-47-18.376278.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/nld_Latn/results_2026-09-28T05-58-55.135152.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/npi_Deva/results_2026-09-28T04-09-06.880792.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/pes_Arab/results_2026-09-28T04-20-12.887711.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/pol_Latn/results_2026-09-28T04-34-11.603940.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/por_Latn/results_2026-09-28T04-48-31.997070.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ron_Latn/results_2026-09-28T05-03-11.387044.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/rus_Cyrl/results_2026-09-28T05-18-45.242782.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/spa_Latn/results_2026-09-28T05-32-57.909369.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/swe_Latn/results_2026-09-28T05-47-19.374812.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/tam_Taml/results_2026-09-28T04-16-18.084980.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/tha_Thai/results_2026-09-28T04-23-21.765065.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/tur_Latn/results_2026-09-28T04-34-16.224462.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/ukr_Cyrl/results_2026-09-28T04-49-06.573758.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/urd_Arab/results_2026-09-28T05-04-31.503175.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/vie_Latn/results_2026-09-28T05-18-19.281027.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/yor_Latn/results_2026-09-28T05-33-13.784791.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/zho_Hans/results_2026-09-28T05-46-49.197011.jsonConfiguration3.0 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/zsm_Latn/results_2026-09-28T05-58-51.198894.jsonConfiguration3.0 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/COMPLETE.jsonConfiguration1.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T213203Z_h013_3628364/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/results_2026-09-29T20-17-45.313600.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/results_2026-09-29T22-52-30.434235.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T025335Z_h013_3634338/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/results_2026-09-30T04-11-17.770656.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/results_2026-09-30T07-30-00.558076.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/results_2026-09-30T10-13-34.425504.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/results_2026-09-30T13-16-45.944774.jsonConfiguration7.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/validation.jsonConfiguration6.3 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ara_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ara_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ben_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ben_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ces_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ces_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_deu_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_deu_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_eng_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_eng_mcf/results_bundle.jsonConfiguration7.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_fra_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_fra_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_hau_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_hau_mcf/results_bundle.jsonConfiguration7.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_heb_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_heb_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ind_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ind_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ita_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ita_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_jpn_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_jpn_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_kor_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_kor_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_msa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_msa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_nep_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_nep_mcf/results_bundle.jsonConfiguration7.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_nld_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_nld_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_pol_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_pol_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_por_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_por_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ron_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ron_mcf/results_bundle.jsonConfiguration7.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_rus_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_rus_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_spa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_spa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_tur_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_tur_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ukr_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_ukr_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_vie_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_vie_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_zho_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/by_task/global_mmlu_zho_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/COMPLETE.jsonConfiguration1.3 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/results_2026-09-28T14-20-09.341872.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/results_2026-09-28T16-51-39.344730.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/results_2026-09-28T19-25-04.067376.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/results_2026-09-28T22-38-00.744746.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/results_2026-09-29T01-22-42.348800.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/results_2026-09-29T04-26-31.053479.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/run_config.jsonConfiguration1.9 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ara_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ara_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ben_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ben_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ces_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ces_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_deu_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_deu_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_eng_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_eng_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_fra_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_fra_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_heb_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_heb_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ind_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ind_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ita_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ita_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_jpn_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_jpn_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_kor_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_kor_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_msa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_msa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_nep_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_nep_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_nld_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_nld_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_pol_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_pol_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_por_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_por_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ron_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ron_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_rus_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_rus_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_spa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_spa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_tur_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_tur_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ukr_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_ukr_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_vie_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_vie_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_yor_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_yor_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_zho_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/by_task/global_mmlu_zho_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/COMPLETE.jsonConfiguration1.4 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/results_2026-09-28T14-24-29.305359.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/results_2026-09-28T16-58-31.365525.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/results_2026-09-28T19-35-20.477270.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260928T233625Z_h020_839639/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T010348Z_h020_983773/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/results_2026-09-29T14-14-49.844975.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/results_2026-09-29T16-55-58.275178.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/COMPLETE.jsonConfiguration1.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/details_manifest.jsonConfiguration2.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/results_2026-09-29T19-57-19.452069.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/run_config.jsonConfiguration2.0 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/validation.jsonConfiguration6.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ara_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ara_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ben_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ben_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ces_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ces_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_deu_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_deu_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_eng_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_eng_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_fra_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_fra_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_heb_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_heb_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ind_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ind_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ita_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ita_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_jpn_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_jpn_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_kor_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_kor_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_msa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_msa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_nep_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_nep_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_nld_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_nld_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_pol_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_pol_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_por_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_por_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ron_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ron_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_rus_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_rus_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_spa_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_spa_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_tur_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_tur_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ukr_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_ukr_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_vie_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_vie_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_yor_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_yor_mcf/results_bundle.jsonConfiguration7.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_zho_mcf/COMPLETE.jsonConfiguration1.1 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/by_task/global_mmlu_zho_mcf/results_bundle.jsonConfiguration7.8 KB —
README.mdDocumentation3.1 KB —
_artifact_manifest/README.mdDocumentation3.1 KB —
finetuning/phase1_lora/top8_expert_lora/README.mdDocumentation1.4 KB —
finetuning/phase1_lora/top8_expert_lora/TMUX_LAUNCH.mdDocumentation3.8 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/README.mdDocumentation5.3 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/README.mdDocumentation4.9 KB —
finetuning/phase2_full/hausa_100k_full/README.mdDocumentation9.5 KB —
finetuning/phase2_full/yoruba_50k_full/README.mdDocumentation4.5 KB —
models/Marco-Nano-Base/README.mdDocumentation7.1 KB —
reproduce/RUN_BELEBELE_34.mdDocumentation80.3 KB —
reproduce/RUN_GLOBAL_MMLU_23.mdDocumentation15.3 KB —
reproduce/RUN_GLOBAL_MMLU_FIXED23_HAU_100K_TOP8.mdDocumentation2.9 KB —
reproduce/RUN_GLOBAL_MMLU_YOR_50K_4GPU.mdDocumentation1.6 KB —
reproduce/global_mmlu_accelerated_3node/README.mdDocumentation10.5 KB —
_artifact_manifest/ALL_SOURCE_FILE_INVENTORY.tsvOther276.2 KB —
_artifact_manifest/BUNDLE_PHYSICAL_SIZE.txtOther114 B —
_artifact_manifest/COPIED_FILES_SHA256.tsvOther42.1 KB —
_artifact_manifest/SIZE_REPORT.tsvOther4.0 KB —
_artifact_manifest/SIZE_SUMMARY.tsvOther565 B —
_artifact_manifest/SOURCES.tsvOther4.3 KB —
_artifact_manifest/UPLOAD_TREE_FILE_INVENTORY.tsvOther195.9 KB —
_artifact_manifest/UPLOAD_TREE_SUMMARY.tsvOther237 B —
_artifact_manifest/VALIDATION_REPORT.txtOther67 B —
finetuning/data/splits/hau_Latn_100k.jsonlOther227.9 MB 86aec98abbf2
finetuning/data/splits/hau_Latn_500k.jsonlOther1.3 GB 4a79cacd3867
finetuning/data/splits/yor_Latn_50k.jsonlOther195.4 MB 6a411053ce4e
finetuning/data/yoruba_100k_fw2_wura/outputs/pre_domain_review_v1/yor_Latn_100k_fw2_wura.jsonlOther389.2 MB 269e19842d27
finetuning/data/yoruba_100k_fw2_wura/outputs/pre_domain_review_v1/yor_Latn_100k_fw2_wura_provenance.parquetOther6.2 MB 903b0cb75468
finetuning/data/yoruba_100k_fw2_wura/outputs/yor_Latn_100k_fw2_wura.jsonlOther388.7 MB f3c55f3c40a5
finetuning/data/yoruba_100k_fw2_wura/outputs/yor_Latn_100k_fw2_wura_provenance.parquetOther6.2 MB 38c4fb8a19af
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/chat_template.jinjaOther2.6 KB —
finetuning/phase1_lora/top8_expert_lora/logs/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn.failed_ddp_find_unused_false_20260930T023031Z.logOther54.3 KB —
finetuning/phase1_lora/top8_expert_lora/logs/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn.logOther120.0 KB —
finetuning/phase1_lora/top8_expert_lora/scripts/nano_yor_50k_pretraining_top8_expert_lora_4gpu.shOther10.7 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/pytorch_model_fsdp_0/__0_0.distcpOther8.2 GB 25cd3aaf03c4
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/pytorch_model_fsdp_0/__1_0.distcpOther8.2 GB 06bc6ea28caf
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/pytorch_model_fsdp_0/__2_0.distcpOther8.2 GB dfa718daa5b3
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/pytorch_model_fsdp_0/__3_0.distcpOther8.2 GB 1eb6917ac0ed
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/chat_template.jinjaOther2.6 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/__pycache__/train_hau_500k_top8.cpython-313.pycOther84.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/logs/05_train_500k_top8_8gpu_20260902T205441Z-pid1747580.logOther365.6 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/logs/05_train_500k_top8_8gpu_20260930T210523Z-pid766975.logOther32.2 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/logs/05_train_500k_top8_8gpu_20260930T212840Z-pid775588.logOther20.2 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/logs/05_train_500k_top8_8gpu_20260930T214251Z-pid781050.logOther27.8 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/logs/05_train_500k_top8_8gpu_20260930T215925Z-pid785986.logOther591.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/scripts/05_train_500k_top8_8gpu.shOther8.3 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_500k_reuse_50k_mask/scripts/06_train_500k_top8_2node_8gpu.shOther14.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/hau_50k/data/matched_hau_Latn_seed42.jsonlOther1.5 MB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/__pycache__/common.cpython-313.pycOther12.9 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/__pycache__/generate_freeze_plan.cpython-313.pycOther19.3 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/__pycache__/validate_checkpoints.cpython-313.pycOther21.8 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/matched_yor_Latn_seed42.jsonlOther1.8 MB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/mode2_outputs/activation_counts_pretraining_corpus_marco_nano_base_yor_yor_Latn_seed42_1lang.progress.jsonlOther7.2 MB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/logs/01_prepare_calibration.logOther1.0 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/logs/02_run_mode2_routing.logOther2.2 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/logs/03_generate_validate_plan.logOther254 B —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/logs/05_train_50k_top8_4gpu_20260926T210504Z.logOther290.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/scripts/01_prepare_calibration.shOther896 B —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/scripts/02_run_mode2_routing.shOther1.4 KB —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/scripts/03_generate_validate_plan.shOther782 B —
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/scripts/05_train_50k_top8_4gpu.shOther10.7 KB —
finetuning/phase2_full/hausa_100k_full/smoke_nano_hau_100k_full_4gpu.shOther13.7 KB —
finetuning/phase2_full/logs/marco_nano_base/hau_Latn_100k_full.logOther587.0 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.dcp_gloo_validation.logOther45.8 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.logOther460.8 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_100k_full.metadata.json.pre-rank-local-optimizer-20261001T172835EDTOther7.5 KB —
finetuning/phase2_full/logs/marco_nano_base/yor_Latn_50k_full.logOther287.8 KB —
finetuning/phase2_full/scripts/nano_hau_100k.shOther19.6 KB —
finetuning/phase2_full/scripts/nano_yor_100k.shOther19.7 KB —
finetuning/phase2_full/scripts/nano_yor_50k.shOther19.6 KB —
finetuning/phase2_full/yoruba_100k_full/finetune_full_time_limited.py.20260930T2135EDT-pre-sharded-checkpoint-fix.bakOther7.2 KB —
finetuning/phase2_full/yoruba_100k_full/finetune_full_time_limited.py.20261001T1650EDT-pre-dcp-gloo.bakOther15.9 KB —
finetuning/phase2_full/yoruba_100k_full/finetune_full_time_limited.py.20261001T1712EDT-pre-rank-local-optimizer.bakOther17.9 KB —
finetuning/phase2_full/yoruba_100k_full/preflight_yoruba_100k_full.py.20260930T2135EDT-pre-sharded-checkpoint-fix.bakOther74.0 KB —
finetuning/phase2_full/yoruba_100k_full/preflight_yoruba_100k_full.py.20261001T1712EDT-pre-rank-local-optimizer.bakOther76.1 KB —
finetuning/phase2_full/yoruba_100k_full/smoke_nano_yor_100k_full_4gpu.shOther13.7 KB —
finetuning/phase2_full/yoruba_100k_full/validate_exp11_dcp_resume.shOther1.6 KB —
finetuning/phase2_full/yoruba_50k_full/smoke_nano_yor_50k_full_4gpu.shOther13.7 KB —
reproduce/global_mmlu_accelerated_3node/run_model_grouped_8gpu.shOther35.5 KB —
reproduce/global_mmlu_fixed23_plus_hau.txtOther744 B —
reproduce/global_mmlu_fixed23_plus_yor.txtOther744 B —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/arb_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/azj_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ben_Beng.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/cat_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ces_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/dan_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/deu_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ell_Grek.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/eng_Latn.logOther7.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/fra_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/hau_Latn.logOther8.7 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/heb_Hebr.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/hun_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ind_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ita_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/jpn_Jpan.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/kaz_Cyrl.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/kor_Hang.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/launcher_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/launcher_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/launcher_shard2.logOther84.2 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/launcher_shard3.logOther123.5 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/nld_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/npi_Deva.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/pes_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/pol_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/por_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ron_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/run_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/run_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/run_shard2.logOther84.2 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/run_shard3.logOther123.5 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/rus_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/spa_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/swe_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/tam_Taml.logOther43.9 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/tha_Thai.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/tur_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/ukr_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/urd_Arab.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/vie_Latn.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/zho_Hans.logOther8.3 KB —
reproduce/results_lighteval_belebele_34/nano_hau_100k_full/_logs/zsm_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/arb_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/azj_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ben_Beng.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/cat_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ces_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/dan_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/deu_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ell_Grek.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/eng_Latn.logOther7.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/fra_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/heb_Hebr.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/hun_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ind_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ita_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/jpn_Jpan.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/kaz_Cyrl.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/kor_Hang.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/launcher_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/launcher_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/launcher_shard2.logOther84.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/launcher_shard3.logOther124.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/nld_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/npi_Deva.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/pes_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/pol_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/por_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ron_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/run_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/run_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/run_shard2.logOther84.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/run_shard3.logOther124.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/rus_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/spa_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/swe_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/tam_Taml.logOther43.9 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/tha_Thai.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/tur_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/ukr_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/urd_Arab.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/vie_Latn.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/yor_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/zho_Hans.logOther8.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_full/_logs/zsm_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/arb_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/azj_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ben_Beng.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/cat_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ces_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/dan_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/deu_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ell_Grek.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/eng_Latn.logOther7.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/fra_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/heb_Hebr.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/hun_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ind_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ita_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/jpn_Jpan.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/kaz_Cyrl.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/kor_Hang.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/launcher_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/launcher_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/launcher_shard2.logOther84.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/launcher_shard3.logOther124.9 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/nld_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/npi_Deva.logOther13.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/pes_Arab.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/pol_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/por_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ron_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/run_shard0.logOther95.6 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/run_shard1.logOther92.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/run_shard2.logOther84.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/run_shard3.logOther124.9 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/rus_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/spa_Latn.logOther9.8 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/swe_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/tam_Taml.logOther43.9 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/tha_Thai.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/tur_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/ukr_Cyrl.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/urd_Arab.logOther13.2 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/vie_Latn.logOther8.5 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/yor_Latn.logOther10.1 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/zho_Hans.logOther8.3 KB —
reproduce/results_lighteval_belebele_34/nano_yor_50k_top8/_logs/zsm_Latn.logOther10.1 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T213203Z_h013_3628364/run.logOther76.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T20-17-45.313600/details_custom|global_mmlu_ara_mcf|5_2026-09-29T20-17-45…Other22.6 MB 3ccb105a7c4b
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T20-17-45.313600/details_custom|global_mmlu_deu_mcf|5_2026-09-29T20-17-45…Other26.6 MB f13a3be472c4
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T20-17-45.313600/details_custom|global_mmlu_spa_mcf|5_2026-09-29T20-17-45…Other25.0 MB 32d97d75805b
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T20-17-45.313600/details_custom|global_mmlu_zho_mcf|5_2026-09-29T20-17-45…Other16.0 MB 2ffd450677b1
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_01/attempts/20260929T214541Z_h013_3634338/run.logOther500.6 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T22-52-30.434235/details_custom|global_mmlu_fra_mcf|5_2026-09-29T22-52-30…Other25.8 MB 9a614ee267d5
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T22-52-30.434235/details_custom|global_mmlu_jpn_mcf|5_2026-09-29T22-52-30…Other24.4 MB 195da3f8bb3f
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T22-52-30.434235/details_custom|global_mmlu_kor_mcf|5_2026-09-29T22-52-30…Other20.5 MB e8a4704acf7f
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-29T22-52-30.434235/details_custom|global_mmlu_por_mcf|5_2026-09-29T22-52-30…Other22.9 MB 1a96f6ab7b1a
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_02/attempts/20260930T001848Z_h013_3634338/run.logOther500.7 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T025335Z_h013_3634338/run.logOther335.0 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T04-11-17.770656/details_custom|global_mmlu_ind_mcf|5_2026-09-30T04-11-17…Other21.8 MB a7cbf13a58b2
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T04-11-17.770656/details_custom|global_mmlu_ita_mcf|5_2026-09-30T04-11-17…Other23.9 MB 02ecf2d52ce1
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T04-11-17.770656/details_custom|global_mmlu_nld_mcf|5_2026-09-30T04-11-17…Other25.3 MB 157a605c4244
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T04-11-17.770656/details_custom|global_mmlu_tur_mcf|5_2026-09-30T04-11-17…Other25.1 MB 6f71b1fb3a5d
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_03/attempts/20260930T053310Z_h013_3775774/run.logOther500.9 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T07-30-00.558076/details_custom|global_mmlu_ben_mcf|5_2026-09-30T07-30-00…Other38.2 MB 62427e23a1c5
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T07-30-00.558076/details_custom|global_mmlu_pol_mcf|5_2026-09-30T07-30-00…Other26.7 MB ba5a8144f89e
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T07-30-00.558076/details_custom|global_mmlu_rus_mcf|5_2026-09-30T07-30-00…Other27.1 MB f8d1057dd5fa
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T07-30-00.558076/details_custom|global_mmlu_vie_mcf|5_2026-09-30T07-30-00…Other19.9 MB 2c11b03f58ca
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_04/attempts/20260930T081223Z_h013_3775774/run.logOther502.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T10-13-34.425504/details_custom|global_mmlu_ces_mcf|5_2026-09-30T10-13-34…Other25.5 MB 584292898809
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T10-13-34.425504/details_custom|global_mmlu_heb_mcf|5_2026-09-30T10-13-34…Other19.9 MB e0d19d1a7a05
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T10-13-34.425504/details_custom|global_mmlu_msa_mcf|5_2026-09-30T10-13-34…Other20.2 MB 1260fb7e66cf
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T10-13-34.425504/details_custom|global_mmlu_ukr_mcf|5_2026-09-30T10-13-34…Other31.2 MB 47b47422fcc8
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_05/attempts/20260930T113209Z_h013_3775774/run.logOther501.5 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T13-16-45.944774/details_custom|global_mmlu_eng_mcf|5_2026-09-30T13-16-45…Other17.7 MB 2bc508c7e763
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T13-16-45.944774/details_custom|global_mmlu_hau_mcf|5_2026-09-30T13-16-45…Other21.2 MB 2144a34a29c4
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T13-16-45.944774/details_custom|global_mmlu_nep_mcf|5_2026-09-30T13-16-45…Other36.4 MB 90737a9ba6a3
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h013/nano_hau_100k_full/2026-09-30T13-16-45.944774/details_custom|global_mmlu_ron_mcf|5_2026-09-30T13-16-45…Other25.8 MB 115281f71a42
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_bundles/bundle_06/attempts/20260930T141505Z_h013_3775774/run.logOther501.6 KB —
reproduce/results_lighteval_global_mmlu_24_hau_8192/nano_hau_100k_full/_logs/run.logOther3.4 MB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T14-20-09.341872/details_custom|global_mmlu_ara_mcf|5_2026-09-28T14-20-09.34…Other22.6 MB 4ac98d8ce36a
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T14-20-09.341872/details_custom|global_mmlu_deu_mcf|5_2026-09-28T14-20-09.34…Other26.6 MB ddb01cfd769e
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T14-20-09.341872/details_custom|global_mmlu_spa_mcf|5_2026-09-28T14-20-09.34…Other25.0 MB b90dcefd0464
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T14-20-09.341872/details_custom|global_mmlu_zho_mcf|5_2026-09-28T14-20-09.34…Other16.0 MB d4bb2d0b5ccb
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_01/attempts/20260928T155035Z_h007_224204/run.logOther513.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T16-51-39.344730/details_custom|global_mmlu_fra_mcf|5_2026-09-28T16-51-39.34…Other25.8 MB f5e401fe85d7
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T16-51-39.344730/details_custom|global_mmlu_jpn_mcf|5_2026-09-28T16-51-39.34…Other24.4 MB eee4cef03452
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T16-51-39.344730/details_custom|global_mmlu_kor_mcf|5_2026-09-28T16-51-39.34…Other20.5 MB cde8e12a3c98
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T16-51-39.344730/details_custom|global_mmlu_por_mcf|5_2026-09-28T16-51-39.34…Other22.9 MB 4db3c9658d5b
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_02/attempts/20260928T182122Z_h007_224204/run.logOther500.6 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T19-25-04.067376/details_custom|global_mmlu_ind_mcf|5_2026-09-28T19-25-04.06…Other21.8 MB 18ad4a4b2601
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T19-25-04.067376/details_custom|global_mmlu_ita_mcf|5_2026-09-28T19-25-04.06…Other23.9 MB 458b293d0598
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T19-25-04.067376/details_custom|global_mmlu_nld_mcf|5_2026-09-28T19-25-04.06…Other25.3 MB eecee9439145
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T19-25-04.067376/details_custom|global_mmlu_tur_mcf|5_2026-09-28T19-25-04.06…Other25.1 MB 040a3a6a739a
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_03/attempts/20260928T205244Z_h007_224204/run.logOther500.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T22-38-00.744746/details_custom|global_mmlu_ben_mcf|5_2026-09-28T22-38-00.74…Other38.2 MB f66e94228ff9
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T22-38-00.744746/details_custom|global_mmlu_pol_mcf|5_2026-09-28T22-38-00.74…Other26.7 MB 2cc39e22ebb9
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T22-38-00.744746/details_custom|global_mmlu_rus_mcf|5_2026-09-28T22-38-00.74…Other27.1 MB 741ac7e90df9
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-28T22-38-00.744746/details_custom|global_mmlu_vie_mcf|5_2026-09-28T22-38-00.74…Other19.9 MB 81b06b40a4ef
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_04/attempts/20260928T232611Z_h007_224204/run.logOther502.4 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T01-22-42.348800/details_custom|global_mmlu_ces_mcf|5_2026-09-29T01-22-42.34…Other25.5 MB 2ec7b922f6ca
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T01-22-42.348800/details_custom|global_mmlu_heb_mcf|5_2026-09-29T01-22-42.34…Other19.9 MB 92711200d384
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T01-22-42.348800/details_custom|global_mmlu_msa_mcf|5_2026-09-29T01-22-42.34…Other20.2 MB a0ac672992b8
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T01-22-42.348800/details_custom|global_mmlu_ukr_mcf|5_2026-09-29T01-22-42.34…Other31.2 MB 04f9110ec57b
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_05/attempts/20260929T024013Z_h007_224204/run.logOther501.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T04-26-31.053479/details_custom|global_mmlu_eng_mcf|5_2026-09-29T04-26-31.05…Other17.6 MB f154566649a6
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T04-26-31.053479/details_custom|global_mmlu_nep_mcf|5_2026-09-29T04-26-31.05…Other38.7 MB d23bb3085604
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T04-26-31.053479/details_custom|global_mmlu_ron_mcf|5_2026-09-29T04-26-31.05…Other24.4 MB 325fc463890f
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h007/nano_yor_50k_full/2026-09-29T04-26-31.053479/details_custom|global_mmlu_yor_mcf|5_2026-09-29T04-26-31.05…Other25.0 MB cd62456791a2
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_bundles/bundle_06/attempts/20260929T052411Z_h007_224204/run.logOther501.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_full/_logs/run.logOther3.0 MB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T14-24-29.305359/details_custom|global_mmlu_ara_mcf|5_2026-09-28T14-24-29.30…Other22.6 MB 4cd1fdd900f1
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T14-24-29.305359/details_custom|global_mmlu_deu_mcf|5_2026-09-28T14-24-29.30…Other26.6 MB de6b54259987
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T14-24-29.305359/details_custom|global_mmlu_spa_mcf|5_2026-09-28T14-24-29.30…Other25.0 MB 1805eb6c7298
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T14-24-29.305359/details_custom|global_mmlu_zho_mcf|5_2026-09-28T14-24-29.30…Other16.0 MB 78de93d685ee
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_01/attempts/20260928T155035Z_h020_839639/run.logOther500.5 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T16-58-31.365525/details_custom|global_mmlu_fra_mcf|5_2026-09-28T16-58-31.36…Other25.8 MB b9215b55ab90
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T16-58-31.365525/details_custom|global_mmlu_jpn_mcf|5_2026-09-28T16-58-31.36…Other24.4 MB 9043a2dc7734
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T16-58-31.365525/details_custom|global_mmlu_kor_mcf|5_2026-09-28T16-58-31.36…Other20.5 MB 769a21fdab67
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T16-58-31.365525/details_custom|global_mmlu_por_mcf|5_2026-09-28T16-58-31.36…Other22.9 MB 449b8a89effd
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_02/attempts/20260928T182532Z_h020_839639/run.logOther500.6 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T19-35-20.477270/details_custom|global_mmlu_ind_mcf|5_2026-09-28T19-35-20.47…Other21.8 MB 39f31acc018b
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T19-35-20.477270/details_custom|global_mmlu_ita_mcf|5_2026-09-28T19-35-20.47…Other23.9 MB aaf9e85757db
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T19-35-20.477270/details_custom|global_mmlu_nld_mcf|5_2026-09-28T19-35-20.47…Other25.3 MB 746d27b95377
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-28T19-35-20.477270/details_custom|global_mmlu_tur_mcf|5_2026-09-28T19-35-20.47…Other25.1 MB e60d20b1580e
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_03/attempts/20260928T205941Z_h020_839639/run.logOther500.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260928T233625Z_h020_839639/run.logOther29.8 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T010348Z_h020_983773/run.logOther431.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T14-14-49.844975/details_custom|global_mmlu_ben_mcf|5_2026-09-29T14-14-49.8…Other38.2 MB 161a9b0649fe
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T14-14-49.844975/details_custom|global_mmlu_pol_mcf|5_2026-09-29T14-14-49.8…Other26.7 MB 72493731dece
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T14-14-49.844975/details_custom|global_mmlu_rus_mcf|5_2026-09-29T14-14-49.8…Other27.1 MB c3a38e21ccbe
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T14-14-49.844975/details_custom|global_mmlu_vie_mcf|5_2026-09-29T14-14-49.8…Other19.9 MB d2c47124678b
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_04/attempts/20260929T145815Z_h020_1229885/run.logOther523.6 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T16-55-58.275178/details_custom|global_mmlu_ces_mcf|5_2026-09-29T16-55-58.2…Other25.5 MB 1b41ba33996b
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T16-55-58.275178/details_custom|global_mmlu_heb_mcf|5_2026-09-29T16-55-58.2…Other19.9 MB 76cfede065b2
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T16-55-58.275178/details_custom|global_mmlu_msa_mcf|5_2026-09-29T16-55-58.2…Other20.2 MB 0a7273e7a474
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T16-55-58.275178/details_custom|global_mmlu_ukr_mcf|5_2026-09-29T16-55-58.2…Other31.2 MB c628f17e4a05
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_05/attempts/20260929T181658Z_h020_1229885/run.logOther502.2 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T19-57-19.452069/details_custom|global_mmlu_eng_mcf|5_2026-09-29T19-57-19.4…Other17.6 MB e18821c2d8bd
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T19-57-19.452069/details_custom|global_mmlu_nep_mcf|5_2026-09-29T19-57-19.4…Other38.7 MB 549dc0033d4c
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T19-57-19.452069/details_custom|global_mmlu_ron_mcf|5_2026-09-29T19-57-19.4…Other24.4 MB aa9309074998
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/details/reproduce/global_mmlu_accelerated_3node/.runtime_models/h020/nano_yor_50k_top8/2026-09-29T19-57-19.452069/details_custom|global_mmlu_yor_mcf|5_2026-09-29T19-57-19.4…Other25.0 MB f93c4f44e57a
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_bundles/bundle_06/attempts/20260929T205742Z_h020_1229885/run.logOther502.3 KB —
reproduce/results_lighteval_global_mmlu_24_yor_8192/nano_yor_50k_top8/_logs/run.logOther3.5 MB —
reproduce/run_lighteval_belebele_34_nano_hau_100k_full_shard0.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_hau_100k_full_shard1.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_hau_100k_full_shard2.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_hau_100k_full_shard3.shOther3.0 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_full_shard0.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_full_shard1.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_full_shard2.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_full_shard3.shOther3.0 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_top8_shard0.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_top8_shard1.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_top8_shard2.shOther2.9 KB —
reproduce/run_lighteval_belebele_34_nano_yor_50k_top8_shard3.shOther3.1 KB —
reproduce/run_lighteval_global_mmlu_nano_hau_100k_full_4gpu.shOther2.0 KB —
reproduce/run_lighteval_global_mmlu_nano_yor_50k_full_4gpu.shOther2.0 KB —
reproduce/run_lighteval_global_mmlu_nano_yor_50k_top8_4gpu.shOther2.0 KB —
requirements_frozen_lighteval_accelerate.txtOther2.7 KB —
.gitattributesRepository5.0 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_100k_full/checkpoint-519/pytorch_model_fsdp_0/.metadataRepository18.7 MB e454562af4fd
finetuning/phase2_full/freeze_experiments/pretraining_corpus_top8/yor_50k/data/.prepare_progress.jsonlRepository1.8 MB —
finetuning/phase2_full/logs/marco_nano_base/.hau_Latn_100k_full.status.json.lockRepository —
finetuning/phase2_full/logs/marco_nano_base/.yor_Latn_100k_full.status.json.lockRepository —
finetuning/phase2_full/logs/marco_nano_base/.yor_Latn_50k_full.status.json.lockRepository —
models/Marco-Nano-Base/.gitattributesRepository1.5 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/merges.txtTokenizer1.7 MB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase1_lora/top8_expert_lora/checkpoints/marco_nano_base/yor_Latn_50k_lora_rank64_pretraining_corpus_top8_experts_no_attn/checkpoint-335/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1500/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/checkpoint-1563/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/tokenizer.jsonTokenizer11.4 MB aeb13307a71a
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_100k_full/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/hau_Latn_500k_exp3_train_only_hau_pretraining_corpus_top8_reuse_50k_mask_experts/checkpoint-1682/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-400/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-500/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-600/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-700/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/checkpoint-782/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/tokenizer.jsonTokenizer7.0 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/tokenizer_config.jsonTokenizer8.1 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_exp3_train_only_yor_pretraining_corpus_top8_experts/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-500/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/checkpoint-782/vocab.jsonTokenizer2.8 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/merges.txtTokenizer1.7 MB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/tokenizer.jsonTokenizer11.4 MB b7a6a993d40b
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/tokenizer_config.jsonTokenizer5.4 KB —
finetuning/phase2_full/checkpoints/marco_nano_base/yor_Latn_50k_full/vocab.jsonTokenizer2.8 MB —
models/Marco-Nano-Base/tokenizer.jsonTokenizer7.0 MB —
models/Marco-Nano-Base/tokenizer_config.jsonTokenizer8.1 KB —
models/Marco-Nano-Base/vocab.jsonTokenizer2.8 MB —

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
1.1 TB
Download from Liang Liang

Released by Liang Liang through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published1.1 TB

Weights only, from the published parameter count; the key-value cache and runtime add to this.