Skip to page content
SAVRN
Search Plan a project

Open-weight model

MoS-8B-Checkpoints

by Ryan Wang ryan-0608/MoS-8B-Checkpoints

MoS-8B-Checkpoints is an open-weight model from Ryan Wang. Its published files total 1.2 TB.

Checkpoints of the Qwen3-8B MoS DFlash draft models (clean-800K E5 follow-up arms, 8B self-generated corpus dataclean-8b-20260912) and the warm-start inputs they were trained from.

Parameters—
Context—
Weights622.6 GB
License—
AccessOpen weights
Monthly Downloads—

Model Card

Checkpoints of the Qwen3-8B MoS DFlash draft models (clean-800K E5 follow-up arms, 8B self-generated corpus dataclean-8b-20260912) and the warm-start inputs they were trained from. Project: MoS (mixture of layer-local expert MLPs added to a shared MLP in each DFlash draft layer). Milestones: 13,325 steps = 1 epoch (E1); 26,650 E2; 39,975 E3; 53,300 E4; 66,625 E5; 6,662 / 19,986 / 33,310 / 46,634 / 59,958 are half-epoch saves. Results, configs and logs: GitHub MoS repository, experiments/aurora/qwen3-8b-800k-e5/bal01/README.md (§10). Archive records (file counts and byte totals, verified against the local copies before deletion): experiments/aurora/qwen3-8b-800k-e5/archive/.

Excerpt from the card by Ryan Wang.

Identity and Version

Repository
ryan-0608/MoS-8B-Checkpoints
Publisher
Ryan Wang
Task
Not stated by the source
Modality
Other
Library
Not stated by the source
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
3b34e273e78c671649974bf6a0e221dce868149b
First published
2026-09-22
Last updated
2026-10-02

Files and Weights

402 files, 1.2 TB in total. The weights are 90 files totalling 622.6 GB in pt, safetensors.

Weights90 files · 622.6 GB
Configuration135 files · 150.4 KB
Documentation1 file · 2.9 KB
Other150 files · 592.8 GB
Repository26 files · 5.1 MB
Every file
FileTypeSizeSHA-256
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/hf/model.safetensorsWeights2.1 GB e0810f8199da
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.7 KB 9fcd38c153dd
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.7 KB 0d5df0906bc4
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.7 KB 7f7d5b028ae2
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.7 KB c75c989c2b9d
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.8 KB eab0c4fc5600
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.8 KB e4b0d079f4be
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights26.3 GB 840c28240292
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights26.3 GB ef1c1a9d30bb
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.8 KB e411ab0f2ff1
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.7 KB e44564351211
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.8 KB cae39b5c53f1
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.8 KB 5822b1ec88f2
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.7 KB 0674cb928093
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.8 KB f0da656b8b5c
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/hf/model.safetensorsWeights26.3 GB 7d7d14bf3aaa
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00000.ptWeights14.8 KB 7d506b5ac3b9
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00001.ptWeights14.7 KB fa024f908983
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00002.ptWeights14.8 KB 4da3ff1d9acb
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00003.ptWeights14.8 KB c10b4aba38f5
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00004.ptWeights14.7 KB 70f3b4ac1353
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/rng_rank_00005.ptWeights14.8 KB 968437143c9e
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights26.3 GB 15f532e747f3
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights26.3 GB 017453b9f46e
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0019986/hf/model.safetensorsWeights26.3 GB aa18eb8204f5
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0026650/hf/model.safetensorsWeights26.3 GB 6a0131692d72
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0033310/hf/model.safetensorsWeights26.3 GB c3cf419e922b
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0039975/hf/model.safetensorsWeights26.3 GB dccbf5e51bbf
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0046634/hf/model.safetensorsWeights26.3 GB 12a8378307dd
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0053300/hf/model.safetensorsWeights26.3 GB bf11e720cfbc
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0059958/hf/model.safetensorsWeights26.3 GB de27e28f7061
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0066625/hf/model.safetensorsWeights26.3 GB 24617e42ffcd
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights5.1 GB abe69ade4f90
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.7 KB 7dd96abdc6ee
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.8 KB f29fa43dd5ae
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.7 KB dc7896276cf2
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.7 KB 97e751831747
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.8 KB 48e3d5b85531
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.8 KB a3ff0d4d4fe3
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/hf/model.safetensorsWeights5.1 GB c736ed46c767
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00000.ptWeights14.7 KB b484502c89d0
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00001.ptWeights14.8 KB 4b9ed05ab5da
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00002.ptWeights14.7 KB 474fd6ed5133
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00003.ptWeights14.7 KB 40fd45801592
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00004.ptWeights14.8 KB 2758fda1634f
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/rng_rank_00005.ptWeights14.8 KB 95be1b0ba374
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights8.1 GB d0e7b883232f
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights8.1 GB 5637a408c386
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.7 KB f5a033eafee8
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.8 KB 740ea40ab0f0
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.7 KB 60a8eb021038
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.8 KB fdc1816ab1d3
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.8 KB 1653d3d59e44
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.6 KB 78884efeece0
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights8.1 GB 014f16116481
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights8.1 GB 2e0e1f97c4cd
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.8 KB 8c71939a3062
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.8 KB 9420a7422385
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.7 KB f9be7ddf2f17
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.8 KB 2aa568cd255f
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.8 KB 9031ad413fd9
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.7 KB 0fa61b087403
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0013325/hf/model.safetensorsWeights8.1 GB 311d3452acfe
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0019986/hf/model.safetensorsWeights8.1 GB 6355f42720cf
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0026650/hf/model.safetensorsWeights8.1 GB 1dab976267b5
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights8.1 GB 4109045502d4
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0026650/hf/model.safetensorsWeights8.1 GB 4d1e9b48a747
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0033310/hf/model.safetensorsWeights8.1 GB ff07979a2fbc
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0039975/hf/model.safetensorsWeights8.1 GB 812cb1d1670a
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0046634/hf/model.safetensorsWeights8.1 GB 47a7f0675e2a
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0053300/hf/model.safetensorsWeights8.1 GB fa25b2aa88c7
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0059958/hf/model.safetensorsWeights8.1 GB de615e0e985d
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0066625/hf/model.safetensorsWeights8.1 GB 72f69d83d132
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0006662/hf/model.safetensorsWeights8.1 GB 3a8d65eaa67e
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/hf/model.safetensorsWeights8.1 GB 8ebfddcb972c
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00000.ptWeights14.8 KB 24ccafc827a0
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00001.ptWeights14.7 KB a457dc47a06c
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00002.ptWeights14.8 KB fabd3cb1e001
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00003.ptWeights14.8 KB 61dfdf7b0cde
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00004.ptWeights14.8 KB 46adb54866bc
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/rng_rank_00005.ptWeights14.8 KB be17faf64632
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0046634/hf/model.safetensorsWeights14.2 GB 7ddc6124defd
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0053300/hf/model.safetensorsWeights14.2 GB 053d82ae51e9
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0059958/hf/model.safetensorsWeights14.2 GB dab338a87f1f
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0066625/hf/model.safetensorsWeights14.2 GB f18ee13fb63c
inputs/d0-fullwidth-init-8b-jitter/model.safetensorsWeights8.1 GB 9653b3e97bdd
inputs/d0-fullwidth-init-8b-x16-jitter/model.safetensorsWeights26.3 GB c3f8adec6717
inputs/d0-shared-random-experts-8b-x16/model.safetensorsWeights26.3 GB 0d7b9448757b
inputs/d0-shared-random-experts-8b-x4-half/model.safetensorsWeights5.1 GB 8d4c4583ba7d
inputs/d0-shared-random-experts-8b-x4/model.safetensorsWeights8.1 GB 6851487ff2a3
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/hf/config.jsonConfiguration1.1 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration544 B —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0019986/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0019986/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0026650/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0026650/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0033310/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0033310/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0039975/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0039975/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0046634/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0046634/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0053300/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0053300/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0059958/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0059958/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0066625/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x16rand-mos/checkpoints/iter_0066625/hf/export_manifest.jsonConfiguration558 B —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration556 B —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration556 B —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0019986/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0019986/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0026650/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos-r2/checkpoints/iter_0026650/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration556 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0026650/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0026650/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0033310/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0033310/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0039975/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0039975/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0046634/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0046634/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0053300/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0053300/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0059958/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0059958/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0066625/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-mos/checkpoints/iter_0066625/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0006662/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0006662/hf/export_manifest.jsonConfiguration556 B —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/meta.jsonConfiguration3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00000.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00001.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00002.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00003.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00004.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/state_fingerprint_rank_00005.jsonConfiguration1.0 KB —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0046634/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0046634/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0053300/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0053300/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0059958/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0059958/hf/export_manifest.jsonConfiguration557 B —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0066625/hf/config.jsonConfiguration1.3 KB —
e5_8b_800k_20260831/arms/e5-x8rand-mos/checkpoints/iter_0066625/hf/export_manifest.jsonConfiguration557 B —
inputs/d0-fullwidth-init-8b-jitter/config.jsonConfiguration1.1 KB —
inputs/d0-fullwidth-init-8b-x16-jitter/config.jsonConfiguration1.1 KB —
inputs/d0-shared-random-experts-8b-x16/config.jsonConfiguration1.1 KB —
inputs/d0-shared-random-experts-8b-x4-half/config.jsonConfiguration1.1 KB —
inputs/d0-shared-random-experts-8b-x4/config.jsonConfiguration1.1 KB —
README.mdDocumentation2.9 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__0_0.distcpOther557.2 MB 1a64447448e7
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__1_0.distcpOther557.2 MB 90765ac2acd4
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__2_0.distcpOther557.2 MB ac77dfecc84f
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__3_0.distcpOther557.2 MB b56d17b00dc1
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__4_0.distcpOther557.2 MB ec9f92d99f2b
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/__5_0.distcpOther556.5 MB 3c003d8a3591
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__0_0.distcpOther2.1 GB de6065c83d26
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__1_0.distcpOther2.1 GB 28b218ee3803
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__2_0.distcpOther2.1 GB 39fb8db0c9a9
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__3_0.distcpOther2.1 GB 024db1433d4a
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__4_0.distcpOther2.1 GB b0c6801f25be
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/__5_0.distcpOther2.1 GB 208442a68bc4
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__0_0.distcpOther9.6 GB 2f0b52568134
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__1_0.distcpOther9.6 GB bb182622050e
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__2_0.distcpOther9.6 GB 36eef91b4053
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__3_0.distcpOther9.6 GB faeb6ead5481
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__4_0.distcpOther6.6 GB d4f300bda40e
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/__5_0.distcpOther6.6 GB 67505bbd8962
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__0_0.distcpOther20.2 GB 3ddc60e07a01
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__1_0.distcpOther20.2 GB 6c49d523aef9
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__2_0.distcpOther20.2 GB 997143772d1f
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__3_0.distcpOther20.2 GB e6b4e4dd567f
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__4_0.distcpOther14.2 GB efc149559288
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/__5_0.distcpOther14.2 GB 92750d485307
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__0_0.distcpOther9.6 GB 0d6934cf58e3
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__1_0.distcpOther9.6 GB 56f629916e98
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__2_0.distcpOther9.6 GB 36d6fcf4d3d1
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__3_0.distcpOther9.6 GB ee618d3008f9
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__4_0.distcpOther6.6 GB 705335dc08a2
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/__5_0.distcpOther6.6 GB b5ac4ffa6854
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__0_0.distcpOther20.2 GB 8852b5693f49
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__1_0.distcpOther20.2 GB 9f1b2976df7c
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__2_0.distcpOther20.2 GB e993ded26f66
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__3_0.distcpOther20.2 GB ed8166bc249b
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__4_0.distcpOther14.2 GB c9d79c264eaa
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/__5_0.distcpOther14.2 GB 684e6269e84b
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__0_0.distcpOther1.1 GB 02f4766676d2
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__1_0.distcpOther1.1 GB 443f2445a459
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__2_0.distcpOther1.1 GB 02b99c473ba1
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__3_0.distcpOther1.1 GB 2950b603b57a
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__4_0.distcpOther1.1 GB b7cfbde548cc
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/__5_0.distcpOther1.1 GB 5c1b85293ea5
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__0_0.distcpOther1.1 GB 73bdb021c7ef
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__1_0.distcpOther1.1 GB eca250675f37
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__2_0.distcpOther1.1 GB 818043d66e6f
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__3_0.distcpOther1.1 GB fad43d836937
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__4_0.distcpOther1.1 GB 4b48362da5f3
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/__5_0.distcpOther1.1 GB 47935609a5c5
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__0_0.distcpOther5.1 GB fe7dbfae5a8f
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__1_0.distcpOther5.1 GB 9bee2339a741
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__2_0.distcpOther5.1 GB 4783dde31e18
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__3_0.distcpOther5.1 GB 7e58d188ef4b
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__4_0.distcpOther5.1 GB a23944f01c14
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/__5_0.distcpOther5.1 GB 09dd3380315c
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__0_0.distcpOther1.1 GB 4dd6f88bc997
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__1_0.distcpOther1.1 GB 6119191ee33b
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__2_0.distcpOther1.1 GB 5a300e59eb6d
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__3_0.distcpOther1.1 GB bd0242e56599
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__4_0.distcpOther1.1 GB fa016d78d24d
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/__5_0.distcpOther1.1 GB 08f6c02f8704
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__0_0.distcpOther5.1 GB b08385ed5114
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__1_0.distcpOther5.1 GB f60c5d4c1338
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__2_0.distcpOther5.1 GB 1484f689b924
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__3_0.distcpOther5.1 GB 4f0acc37a844
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__4_0.distcpOther5.1 GB f9179f12a1c2
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/__5_0.distcpOther5.1 GB 565e4d201150
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__0_0.distcpOther1.6 GB 68709225c903
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__1_0.distcpOther1.6 GB 932e1716a3c6
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__2_0.distcpOther1.6 GB edfab4f7f665
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__3_0.distcpOther1.6 GB aaf56ab54e6b
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__4_0.distcpOther1.6 GB a338ca62b0fa
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/__5_0.distcpOther1.6 GB e9f56c2d21ad
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__0_0.distcpOther8.1 GB 9f30b99390a6
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__1_0.distcpOther8.1 GB b813458f1511
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__2_0.distcpOther8.1 GB 8705690a095a
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__3_0.distcpOther8.1 GB c993adb1e399
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__4_0.distcpOther8.1 GB f00cff4e8e5f
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/__5_0.distcpOther8.1 GB 42dce2bb4126
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__0_0.distcpOther1.6 GB f077a66fb91a
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__1_0.distcpOther1.6 GB eeffd85c0a71
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__2_0.distcpOther1.6 GB f07c056a5c6c
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__3_0.distcpOther1.6 GB 620fb9df62f9
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__4_0.distcpOther1.6 GB d49e75631bd2
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/__5_0.distcpOther1.6 GB 388fd9a0842b
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__0_0.distcpOther8.1 GB c6d6e8d164bc
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__1_0.distcpOther8.1 GB 2c18ad7b7efb
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__2_0.distcpOther8.1 GB 204e4acef7c0
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__3_0.distcpOther8.1 GB d17411a44b99
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__4_0.distcpOther8.1 GB 86c7977727dd
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/__5_0.distcpOther8.1 GB 4efc1f274799
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__0_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__1_0.distcpOther3.8 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__2_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__3_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__4_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/__5_0.distcpOther2.5 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__0_0.distcpOther1.6 GB f7bdb9592314
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__1_0.distcpOther1.6 GB b14d5ed70b9e
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__2_0.distcpOther1.6 GB a94f1908ed75
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__3_0.distcpOther1.6 GB cea854d16ec8
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__4_0.distcpOther1.6 GB fd6bc9d39eda
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/__5_0.distcpOther1.6 GB 88ce9a07dd3e
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__0_0.distcpOther8.1 GB 6ce0a1046587
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__1_0.distcpOther8.1 GB e2292dbf2db7
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__2_0.distcpOther8.1 GB 223fb4e2e6d5
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__3_0.distcpOther8.1 GB c4863cb676a2
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__4_0.distcpOther8.1 GB 09673336edda
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/__5_0.distcpOther8.1 GB 7aa3b46f8d8e
.gitattributesRepository16.2 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/model/.metadataRepository80.4 KB —
e5_8b_800k_20260831/arms/e5-dense-lr6e4/checkpoints/iter_0013325/optimizer/.metadataRepository209.0 KB bed800749594
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/model/.metadataRepository189.2 KB fcc89bfc96d9
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0013325/optimizer/.metadataRepository507.3 KB 865567be4d18
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/model/.metadataRepository189.2 KB 15e2c6c8e1d7
e5_8b_800k_20260831/arms/e5-x16lr4-mos/checkpoints/iter_0026650/optimizer/.metadataRepository507.3 KB 84c6beb03458
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0006662/model/.metadataRepository179.7 KB 24cc624d860e
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/model/.metadataRepository179.7 KB 3929f812ae7e
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0013325/optimizer/.metadataRepository468.0 KB 504169eaf96c
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/model/.metadataRepository179.7 KB 64b158cea570
e5_8b_800k_20260831/arms/e5-x4rand-half-mos/checkpoints/iter_0026650/optimizer/.metadataRepository468.0 KB 981c84dd8b24
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/model/.metadataRepository179.7 KB e5d7c5c81ce6
e5_8b_800k_20260831/arms/e5-x4rand-lr1e3-mos/checkpoints/iter_0013325/optimizer/.metadataRepository468.7 KB e3a5c3856825
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/model/.metadataRepository179.7 KB 29a3d022daf8
e5_8b_800k_20260831/arms/e5-x4rand-lr1e4-mos/checkpoints/iter_0013325/optimizer/.metadataRepository464.6 KB 8ccee613d255
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/lr_scheduler/.metadataRepository3.4 KB —
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/model/.metadataRepository179.7 KB ac8738e1d369
e5_8b_800k_20260831/arms/e5-x4rand-top1-mos/checkpoints/iter_0013325/optimizer/.metadataRepository468.7 KB 74641aaee7cd

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
622.6 GB
Download from Ryan Wang

Released by Ryan Wang through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published622.6 GB

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