effpretrain_ckpts is an open-weight model from Dmitrii Andriianov. Its published files total 701.2 GB.
Parameters—
Context—
Weights699.4 GB
License—
AccessOpen weights
Monthly Downloads—
Model Card
The publisher has not written a card for this model.
Identity and Version
- Repository
- DimaTivator/effpretrain_ckpts
- Publisher
- Dmitrii Andriianov
- 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
- c6ba97c151f4190c0b106d7dd55f4c1643467405
- First published
- 2026-06-27
- Last updated
- 2026-10-13
Files and Weights
785 files, 701.2 GB in total. The weights are 392 files totalling 699.4 GB in npz, pt.
Weights392 files · 699.4 GB
Configuration189 files · 78.4 MB
Documentation1 file · 1.7 KB
Other200 files · 1.7 GB
Repository3 files · 1.8 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| 1xChinchilla_fp8_resume/adamw_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 3.1 GB | 4d0bf08ad92c |
| 1xChinchilla_fp8_resume/adamw_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | a3f9df099f6a |
| 1xChinchilla_fp8_resume/ademamix_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 4.1 GB | 7466e214e815 |
| 1xChinchilla_fp8_resume/ademamix_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | a3f9df099f6a |
| 1xChinchilla_fp8_resume/muon_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | d98c1cb09497 |
| 1xChinchilla_fp8_resume/muon_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | a3f9df099f6a |
| 1xChinchilla_fp8_resume/soap_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 6.5 GB | 208bc09098a9 |
| 1xChinchilla_fp8_resume/soap_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | a3f9df099f6a |
| 1xChinchilla_full_fp8_resume/500m_adamw_full_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 5.9 GB | 20ccd9ce3ecb |
| 1xChinchilla_full_fp8_resume/500m_adamw_full_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 130.7 KB | d92c4ea1474d |
| 1xChinchilla_full_fp8_resume/500m_ademamix_full_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 7.9 GB | da4c82e7ee39 |
| 1xChinchilla_full_fp8_resume/500m_ademamix_full_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 130.7 KB | d92c4ea1474d |
| 1xChinchilla_full_fp8_resume/500m_muon_full_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 4.5 GB | 0458b765a8e8 |
| 1xChinchilla_full_fp8_resume/500m_muon_full_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 130.7 KB | d92c4ea1474d |
| 1xChinchilla_full_fp8_resume/500m_soap_full_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 14.1 GB | 54a479e3db92 |
| 1xChinchilla_full_fp8_resume/500m_soap_full_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 130.7 KB | d92c4ea1474d |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/67911/main.pt | Weights | 3.1 GB | e2246f976f62 |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/67911/worker_0.pt | Weights | 273.5 KB | 11bbd213cc52 |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/67911/worker_1.pt | Weights | 163.9 KB | 55b200ddf2ee |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/latest/main.pt | Weights | 3.1 GB | ff7f8d6782e8 |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/latest/worker_0.pt | Weights | 128.0 KB | f6e794165885 |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/ckpts/latest/worker_1.pt | Weights | 142.4 KB | ac0ee3981752 |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/67911/main.pt | Weights | 2.7 GB | d8fa8706faa4 |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/67911/worker_0.pt | Weights | 273.5 KB | 20a48297b0c6 |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/67911/worker_1.pt | Weights | 163.9 KB | 6fd8b2f3bba6 |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/latest/main.pt | Weights | 2.7 GB | c33040dec2f3 |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/latest/worker_0.pt | Weights | 101.0 KB | bd2751dc8b3f |
| 1xChinchilla_optimizer_fp8/500m_muon_optimizer_fp8_1xC/ckpts/latest/worker_1.pt | Weights | 89.1 KB | 0adfccbda431 |
| 1xChinchilla_optimizer_fp8/500m_soap_optimizer_fp8_1xC_bs16_acc8/ckpts/67911/main.pt | Weights | 11.2 GB | 5046aed5778d |
| 1xChinchilla_optimizer_fp8/500m_soap_optimizer_fp8_1xC_bs16_acc8/ckpts/67911/worker_0.pt | Weights | 179.4 KB | 8b4ea122089c |
| 1xChinchilla_optimizer_fp8/500m_soap_optimizer_fp8_1xC_bs16_acc8/ckpts/latest/main.pt | Weights | 11.2 GB | e3fd4d7e9757 |
| 1xChinchilla_optimizer_fp8/500m_soap_optimizer_fp8_1xC_bs16_acc8/ckpts/latest/worker_0.pt | Weights | 130.7 KB | 79f8a02e73ae |
| 1xChinchilla_resume_1gpu/lion_muon_lr1e-3_1xC_resume_1gpu/ckpts/latest/main.pt | Weights | 4.5 GB | 4ba32d55ef87 |
| 1xChinchilla_resume_1gpu/lion_muon_lr1e-3_1xC_resume_1gpu/ckpts/latest/worker_0.pt | Weights | 130.7 KB | d92c4ea1474d |
| 2xChinchilla_1gpu/lion_muon_lr1e-3_2xC_1gpu/ckpts/67911/main.pt | Weights | 4.5 GB | b6abe80fd6fb |
| 2xChinchilla_1gpu/lion_muon_lr1e-3_2xC_1gpu/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_1gpu/lion_muon_lr1e-3_2xC_1gpu/ckpts/latest/main.pt | Weights | 4.5 GB | a166390a7990 |
| 2xChinchilla_1gpu/lion_muon_lr1e-3_2xC_1gpu/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_fp8/500m_adamw_act_fp8_2xC/ckpts/67911/main.pt | Weights | 5.9 GB | 78db70b2b5fd |
| 2xChinchilla_fp8/500m_adamw_act_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_fp8/500m_adamw_act_fp8_2xC/ckpts/latest/main.pt | Weights | 5.9 GB | d2b6bab9a563 |
| 2xChinchilla_fp8/500m_adamw_act_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_fp8/500m_ademamix_act_fp8_2xC/ckpts/67911/main.pt | Weights | 7.9 GB | 3c80e1867832 |
| 2xChinchilla_fp8/500m_ademamix_act_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_fp8/500m_ademamix_act_fp8_2xC/ckpts/latest/main.pt | Weights | 7.9 GB | 635882fbb295 |
| 2xChinchilla_fp8/500m_ademamix_act_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_fp8/500m_muon_act_fp8_2xC/ckpts/67911/main.pt | Weights | 4.5 GB | 01e93471dcd8 |
| 2xChinchilla_fp8/500m_muon_act_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_fp8/500m_muon_act_fp8_2xC/ckpts/latest/main.pt | Weights | 4.5 GB | 81e3bc638e21 |
| 2xChinchilla_fp8/500m_muon_act_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_fp8_2gpus/adamw_act_fp8_2xC_2gpus/ckpts/latest/main.pt | Weights | 5.9 GB | 8f551061e8c2 |
| 2xChinchilla_fp8_2gpus/adamw_act_fp8_2xC_2gpus/ckpts/latest/worker_0.pt | Weights | 161.5 KB | 9d3eb6d53ac0 |
| 2xChinchilla_fp8_2gpus/adamw_act_fp8_2xC_2gpus/ckpts/latest/worker_1.pt | Weights | 98.9 KB | a84295025e49 |
| 2xChinchilla_fp8_2gpus/ademamix_act_fp8_2xC_2gpus/ckpts/latest/main.pt | Weights | 7.9 GB | 7f3d193f39bc |
| 2xChinchilla_fp8_2gpus/ademamix_act_fp8_2xC_2gpus/ckpts/latest/worker_0.pt | Weights | 163.6 KB | 8eecfbeb0c42 |
| 2xChinchilla_fp8_2gpus/ademamix_act_fp8_2xC_2gpus/ckpts/latest/worker_1.pt | Weights | 17.9 KB | f114e8382432 |
| 2xChinchilla_fp8_resume/adamw_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 3.1 GB | 71909b563764 |
| 2xChinchilla_fp8_resume/adamw_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | 4ca567b5d016 |
| 2xChinchilla_fp8_resume/ademamix_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 4.1 GB | a9a5619e9a67 |
| 2xChinchilla_fp8_resume/ademamix_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | 4ca567b5d016 |
| 2xChinchilla_fp8_resume/muon_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | c025608dad04 |
| 2xChinchilla_fp8_resume/muon_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | 4ca567b5d016 |
| 2xChinchilla_fp8_resume/soap_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 6.5 GB | e463fb91a6b1 |
| 2xChinchilla_fp8_resume/soap_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | 4ca567b5d016 |
| 2xChinchilla_full_fp8/500m_adamw_full_fp8_2xC/ckpts/67911/main.pt | Weights | 5.9 GB | 78db70b2b5fd |
| 2xChinchilla_full_fp8/500m_adamw_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_full_fp8/500m_adamw_full_fp8_2xC/ckpts/latest/main.pt | Weights | 5.9 GB | d2b6bab9a563 |
| 2xChinchilla_full_fp8/500m_adamw_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_full_fp8/500m_ademamix_full_fp8_2xC/ckpts/67911/main.pt | Weights | 7.9 GB | 3c80e1867832 |
| 2xChinchilla_full_fp8/500m_ademamix_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_full_fp8/500m_ademamix_full_fp8_2xC/ckpts/latest/main.pt | Weights | 7.9 GB | 635882fbb295 |
| 2xChinchilla_full_fp8/500m_ademamix_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_full_fp8/500m_apollo_full_fp8_2xC/ckpts/67911/main.pt | Weights | 3.9 GB | f6bb9c58e152 |
| 2xChinchilla_full_fp8/500m_apollo_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | 7d264e100b59 |
| 2xChinchilla_full_fp8/500m_apollo_full_fp8_2xC/ckpts/latest/main.pt | Weights | 3.9 GB | 8e43f970ac8e |
| 2xChinchilla_full_fp8/500m_apollo_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ed232f19d2e2 |
| 2xChinchilla_full_fp8/500m_frugal_full_fp8_2xC/ckpts/67911/main.pt | Weights | 3.7 GB | 82b07dc6c135 |
| 2xChinchilla_full_fp8/500m_frugal_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | ca54448abb6c |
| 2xChinchilla_full_fp8/500m_frugal_full_fp8_2xC/ckpts/latest/main.pt | Weights | 3.7 GB | 74c0d9e2f672 |
| 2xChinchilla_full_fp8/500m_frugal_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | 96f82d18c49f |
| 2xChinchilla_full_fp8/500m_galore_full_fp8_2xC/ckpts/67911/main.pt | Weights | 4.6 GB | fbe11804eeae |
| 2xChinchilla_full_fp8/500m_galore_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | 7d264e100b59 |
| 2xChinchilla_full_fp8/500m_galore_full_fp8_2xC/ckpts/latest/main.pt | Weights | 4.6 GB | b35020587025 |
| 2xChinchilla_full_fp8/500m_galore_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ed232f19d2e2 |
| 2xChinchilla_full_fp8/500m_muon_full_fp8_2xC/ckpts/67911/main.pt | Weights | 4.5 GB | 01e93471dcd8 |
| 2xChinchilla_full_fp8/500m_muon_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_full_fp8/500m_muon_full_fp8_2xC/ckpts/latest/main.pt | Weights | 4.5 GB | 81e3bc638e21 |
| 2xChinchilla_full_fp8/500m_muon_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 2xChinchilla_full_fp8/500m_slim_adam_full_fp8_2xC/ckpts/latest/main.pt | Weights | 5.4 GB | c4d8029e6e48 |
| 2xChinchilla_full_fp8/500m_slim_adam_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 172.8 KB | af004346d397 |
| 2xChinchilla_full_fp8/500m_soap_full_fp8_2xC/ckpts/67911/main.pt | Weights | 14.1 GB | 022b611b61fa |
| 2xChinchilla_full_fp8/500m_soap_full_fp8_2xC/ckpts/67911/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| 2xChinchilla_full_fp8/500m_soap_full_fp8_2xC/ckpts/latest/main.pt | Weights | 14.1 GB | b74e8574157b |
| 2xChinchilla_full_fp8/500m_soap_full_fp8_2xC/ckpts/latest/worker_0.pt | Weights | 204.5 KB | ba3b6ddf63ca |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/105975/main.pt | Weights | 2.5 GB | 04b985697f0c |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 17be6d05222a |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/35325/main.pt | Weights | 2.5 GB | 45669aaa7c86 |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | e528453958d5 |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/70650/main.pt | Weights | 2.5 GB | 496a4d6646a4 |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 4ef4672135a3 |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/latest/main.pt | Weights | 2.5 GB | cb78704386d3 |
| 4xChinchilla/muon_lr1e-3_4xC/ckpts/latest/worker_0.pt | Weights | 76.1 KB | 551da86e114d |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/105975/main.pt | Weights | 3.1 GB | 88a7affea9bd |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 17be6d05222a |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/35325/main.pt | Weights | 3.1 GB | 83ca02e5e5d7 |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | e528453958d5 |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/70650/main.pt | Weights | 3.1 GB | c265d7feaf11 |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 4ef4672135a3 |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/latest/main.pt | Weights | 3.1 GB | 30c60e982bc7 |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 86f0fa250678 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/105975/main.pt | Weights | 4.1 GB | 2b1a9e4dc3eb |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 17be6d05222a |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/35325/main.pt | Weights | 4.1 GB | 839bb768f3a5 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | e528453958d5 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/70650/main.pt | Weights | 4.1 GB | 0cc0737b0d85 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 4ef4672135a3 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/latest/main.pt | Weights | 4.1 GB | 7c3d749cb940 |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 86f0fa250678 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.5 GB | b19f258e8f43 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 17be6d05222a |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.5 GB | df776647ba66 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | e528453958d5 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.5 GB | 246d03b95525 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 4ef4672135a3 |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.5 GB | acf9710e0a0e |
| 4xChinchilla_fp8/muon_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 86f0fa250678 |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/105975/main.pt | Weights | 6.5 GB | bcc9c0f66e11 |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 17be6d05222a |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/35325/main.pt | Weights | 6.5 GB | 7e29ce168cbf |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | e528453958d5 |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/70650/main.pt | Weights | 6.5 GB | b21d09a13d7e |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 4ef4672135a3 |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/latest/main.pt | Weights | 6.5 GB | 965d4f6f9dc6 |
| 4xChinchilla_fp8/soap_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 86f0fa250678 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_bf16_wd1e-4_2xC_warmup2000_2gpu-iter-67911/main.pt | Weights | 5.4 GB | edc1feda0f29 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_bf16_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_0.pt | Weights | 14.8 KB | e804fc858e24 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_bf16_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_1.pt | Weights | 14.8 KB | f1a05dfe3822 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_act_wd1e-4_2xC_warmup2000_2gpu-iter-67911/main.pt | Weights | 5.4 GB | 6f1cf75bdacd |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_act_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_0.pt | Weights | 14.8 KB | e804fc858e24 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_act_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_1.pt | Weights | 14.8 KB | f1a05dfe3822 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_full_wd1e-4_2xC_warmup2000_2gpu-iter-67911/main.pt | Weights | 2.9 GB | dd90c151c4eb |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_full_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_0.pt | Weights | 14.8 KB | e804fc858e24 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_full_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_1.pt | Weights | 14.8 KB | f1a05dfe3822 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_w16a16g32_fp8_states_wd1e-4_2xC_warmup2000_2gpu-iter-67911/main.pt | Weights | 2.9 GB | 46dd12e9bcbe |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_w16a16g32_fp8_states_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_0.pt | Weights | 14.8 KB | e804fc858e24 |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_w16a16g32_fp8_states_wd1e-4_2xC_warmup2000_2gpu-iter-67911/worker_1.pt | Weights | 14.8 KB | f1a05dfe3822 |
| analysis/bootstrap_downstream_metrics_20260828/adamw_fp8_act/bootstrap_distributions.npz | Weights | 337.6 KB | 42c5c8fffc58 |
| analysis/bootstrap_downstream_metrics_20260828/adamw_fp8_states/bootstrap_distributions.npz | Weights | 335.6 KB | 45477b0dcb5f |
| analysis/bootstrap_downstream_metrics_20260828/ademamix_fp8_act/bootstrap_distributions.npz | Weights | 318.8 KB | b2d06a9e409e |
| analysis/bootstrap_downstream_metrics_20260828/muon_fp8_act/bootstrap_distributions.npz | Weights | 340.7 KB | 8f33cf74b8be |
| analysis/bootstrap_downstream_metrics_20260828/muon_fp8_states/bootstrap_distributions.npz | Weights | 318.0 KB | 3e25748de827 |
| analysis/bootstrap_downstream_metrics_20260828/soap_fp8_states/bootstrap_distributions.npz | Weights | 1.2 MB | fa04ec8c3257 |
| analysis/bootstrap_downstream_metrics_20260828/soap_report/bootstrap_distributions.npz | Weights | 339.9 KB | 3e2619139e66 |
| analysis/bootstrap_downstream_metrics_20260828_invalid_wiki_batch128/soap_fp8_states/bootstrap_distributions.npz | Weights | 1.2 MB | 11b293e4552a |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/checkpoints/bf16_75457/main.pt | Weights | 4.5 GB | 39313fdd03e9 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/checkpoints/bf16_75457/worker_0.pt | Weights | 63.0 KB | 3c9a16c2eb19 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/checkpoints/bf16_75457/worker_1.pt | Weights | 185.5 KB | 449ca5bdbb05 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/checkpoints/bf16_75457/worker_2.pt | Weights | 69.9 KB | 5ae0eaef5c62 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/checkpoints/bf16_75457/worker_3.pt | Weights | 32.7 KB | fa94985f60aa |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/bootstrap_distributions.npz | Weights | 3.2 MB | 8abd3d16e733 |
| analysis/hf_stage3_dense_upload/adamw_fp8_act/main.pt | Weights | 5.9 GB | a66cab09514a |
| analysis/hf_stage3_dense_upload/adamw_fp8_act/pre_decay/main.pt | Weights | 5.9 GB | 78db70b2b5fd |
| analysis/hf_stage3_dense_upload/adamw_fp8_act/pre_decay/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| analysis/hf_stage3_dense_upload/adamw_fp8_act/worker_0.pt | Weights | 187.5 KB | a43ac6cb2721 |
| analysis/hf_stage3_dense_upload/adamw_fp8_states/main.pt | Weights | 3.1 GB | ff7f8d6782e8 |
| analysis/hf_stage3_dense_upload/adamw_fp8_states/pre_decay/main.pt | Weights | 3.1 GB | e2246f976f62 |
| analysis/hf_stage3_dense_upload/adamw_fp8_states/pre_decay/worker_0.pt | Weights | 273.5 KB | 11bbd213cc52 |
| analysis/hf_stage3_dense_upload/adamw_fp8_states/worker_0.pt | Weights | 128.0 KB | f6e794165885 |
| analysis/hf_stage3_dense_upload/ademamix_fp8_act/main.pt | Weights | 7.9 GB | 9eb040c2b616 |
| analysis/hf_stage3_dense_upload/ademamix_fp8_act/pre_decay/main.pt | Weights | 7.9 GB | 3c80e1867832 |
| analysis/hf_stage3_dense_upload/ademamix_fp8_act/pre_decay/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| analysis/hf_stage3_dense_upload/ademamix_fp8_act/worker_0.pt | Weights | 20.8 KB | 8941ae1b8a06 |
| analysis/hf_stage3_dense_upload/muon_fp8_act/main.pt | Weights | 4.5 GB | 72b254fa04ca |
| analysis/hf_stage3_dense_upload/muon_fp8_act/pre_decay/main.pt | Weights | 4.5 GB | 01e93471dcd8 |
| analysis/hf_stage3_dense_upload/muon_fp8_act/pre_decay/worker_0.pt | Weights | 179.4 KB | dbbcf1743d22 |
| analysis/hf_stage3_dense_upload/muon_fp8_act/worker_0.pt | Weights | 148.7 KB | 13d7d7e9f10e |
| analysis/hf_stage3_dense_upload/muon_fp8_states/main.pt | Weights | 2.7 GB | f75e038321ce |
| analysis/hf_stage3_dense_upload/muon_fp8_states/pre_decay/main.pt | Weights | 2.7 GB | 1567e6279a5a |
| analysis/hf_stage3_dense_upload/muon_fp8_states/pre_decay/worker_0.pt | Weights | 422.3 KB | 5879a4b56ad1 |
| analysis/hf_stage3_dense_upload/muon_fp8_states/worker_0.pt | Weights | 255.3 KB | 274f55b44f57 |
| analysis/hf_stage3_dense_upload/soap_fp8_states/main.pt | Weights | 11.2 GB | b0add45e4e81 |
| analysis/hf_stage3_dense_upload/soap_fp8_states/pre_decay/main.pt | Weights | 11.2 GB | 221f051c78da |
| analysis/hf_stage3_dense_upload/soap_fp8_states/pre_decay/worker_0.pt | Weights | 15.0 KB | ab1b4493da19 |
| analysis/hf_stage3_dense_upload/soap_fp8_states/worker_0.pt | Weights | 15.0 KB | b3c06656523a |
| analysis/report_rebootstrap_20260828/muon_fp8_act_itr77168/bootstrap_distributions.npz | Weights | 1.2 MB | 2f9e74c5d9fc |
| analysis/report_rebootstrap_20260828/soap_fp8_states_seed2/bootstrap_distributions.npz | Weights | 1.2 MB | 114c2c6f0f74 |
| analysis/report_wikitext_ci_20260828/adamw_fp8_act/bootstrap_distributions.npz | Weights | 542.8 KB | 0e77280f09f5 |
| analysis/report_wikitext_ci_20260828/adamw_fp8_states/bootstrap_distributions.npz | Weights | 541.8 KB | 9ee1ab6c0b4c |
| analysis/report_wikitext_ci_20260828/ademamix_fp8_act/bootstrap_distributions.npz | Weights | 542.3 KB | 5aee29c2e79d |
| analysis/report_wikitext_ci_20260828/muon_fp8_states/bootstrap_distributions.npz | Weights | 542.8 KB | c4568d0a761d |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 7944d09cd29c |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_20260816/main.pt | Weights | 494.4 MB | 60cd4afce311 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_causal_124m_paper500_lambda0p1_20260820/prehab_sum/main.pt | Weights | 494.4 MB | 6cca60ff6561 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_causal_124m_paper500_lambda0p1_20260820/task_only/main.pt | Weights | 494.4 MB | c859ab02e5ba |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_causal_20260817/prehab_sum/main.pt | Weights | 494.4 MB | 8ad7cc4300b3 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_causal_20260817/task_only/main.pt | Weights | 494.4 MB | 5ecbc2f85fa8 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | e364e8060770 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 81e3fce0aef0 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 63a6f42a4ac8 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 64f1d59e675a |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 473968b65058 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 398a563f7d15 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 24fc67f329a3 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | db35bb88b5da |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 17940b1be991 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 4516f7cb3bc3 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | b25579862c43 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | f8ef72580a61 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 5726201ab696 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | 7d29f9c0ddf6 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | ea93d5a5bc04 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | b3cf0c112cfb |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 81a1e7d7b72e |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 25c0e2d03fed |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_5_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank/ckpts/latest/main.pt | Weights | 1.5 GB | 60240a9f0edf |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | a82b3e349b75 |
| cf_bruteforce_124M/llama124m_adamw_wd0.1_lr5e-3_finewebedu_rankprofile_20261012/ckpts/latest/main.pt | Weights | 1.5 GB | 32730f167b50 |
| cf_bruteforce_124M/llama124m_adamw_wd0.5_lr5e-3_finewebedu_rankprofile_20261012/ckpts/latest/main.pt | Weights | 1.5 GB | a059bdb1c21e |
| cf_bruteforce_124M/llama124m_adamw_wd1_lr5e-3_finewebedu_rankprofile_20261012/ckpts/latest/main.pt | Weights | 1.5 GB | b4215b9eab29 |
| cf_bruteforce_124M/llama124m_adamw_wd2_lr5e-3_finewebedu_rankprofile_20261012/ckpts/latest/main.pt | Weights | 1.5 GB | ca6369ad2ea5 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 08c022ba6665 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank/ckpts/latest/prehab_20260816/main.pt | Weights | 1.0 GB | ff0a943b7d39 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | ccf278e95905 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | a4a8bea52382 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | edc9969244b6 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 3862fa41ead5 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | df594581f2af |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | e46684f0f247 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 1983d9cb8481 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 3736ba311143 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | b9031a03aa31 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | baab7a8a5b12 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | bd319b7f75f4 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 73da32c18f6e |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 181c5c5e1429 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | eb5c04c18abc |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | d8736f845011 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/ckpts/latest/main.pt | Weights | 3.1 GB | 32a6cb20196f |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | ca56505328e7 |
| efficient_1xChinchilla_fp8_resume/adam_mini_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.1 GB | bf9eb32a79b7 |
| efficient_1xChinchilla_fp8_resume/adam_mini_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_1xChinchilla_fp8_resume/apollo_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | ef0f76e7617b |
| efficient_1xChinchilla_fp8_resume/apollo_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_1xChinchilla_fp8_resume/apollo_mini_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | 5bf5cfc3cb4a |
| efficient_1xChinchilla_fp8_resume/apollo_mini_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_1xChinchilla_fp8_resume/fira_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | e2e04181f0fa |
| efficient_1xChinchilla_fp8_resume/fira_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_1xChinchilla_fp8_resume/frugal_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | 3138a97c5aa4 |
| efficient_1xChinchilla_fp8_resume/frugal_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 51eec0f84b2d |
| efficient_1xChinchilla_fp8_resume/galore_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | 336148d203af |
| efficient_1xChinchilla_fp8_resume/galore_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_1xChinchilla_fp8_resume/slim_adam_act_fp8_1xC_resume/ckpts/latest/main.pt | Weights | 2.7 GB | 0451505ed5f8 |
| efficient_1xChinchilla_fp8_resume/slim_adam_act_fp8_1xC_resume/ckpts/latest/worker_0.pt | Weights | 70.0 KB | 2f24321f35fd |
| efficient_2xChinchilla_fp8_resume/adam_mini_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.1 GB | 474be78634db |
| efficient_2xChinchilla_fp8_resume/adam_mini_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_2xChinchilla_fp8_resume/apollo_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | a0b8fd50709b |
| efficient_2xChinchilla_fp8_resume/apollo_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_2xChinchilla_fp8_resume/apollo_mini_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | b79ebe8303d4 |
| efficient_2xChinchilla_fp8_resume/apollo_mini_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_2xChinchilla_fp8_resume/fira_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | eab378f4b430 |
| efficient_2xChinchilla_fp8_resume/fira_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_2xChinchilla_fp8_resume/frugal_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.2 GB | d3b6cb128506 |
| efficient_2xChinchilla_fp8_resume/frugal_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | 78ac66b5034e |
| efficient_2xChinchilla_fp8_resume/galore_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.5 GB | 7e220a468772 |
| efficient_2xChinchilla_fp8_resume/galore_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_2xChinchilla_fp8_resume/slim_adam_act_fp8_2xC_resume/ckpts/latest/main.pt | Weights | 2.7 GB | a6253c59f86b |
| efficient_2xChinchilla_fp8_resume/slim_adam_act_fp8_2xC_resume/ckpts/latest/worker_0.pt | Weights | 18.1 KB | ba31fb6ea8da |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.1 GB | abb34cb662bd |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.1 GB | c1e534d10d2d |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.1 GB | 6d633291806b |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.1 GB | d955ff7ab9ea |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.2 GB | e8b56e494af1 |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.2 GB | 704a8ad7627e |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.2 GB | beea9284081d |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.2 GB | a6d876ce3c99 |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.2 GB | 0c9a7ecde369 |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.2 GB | 18087a585e4f |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.2 GB | 18a9c5467d10 |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.2 GB | 56143300d881 |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| efficient_4xChinchilla_fp8/badam_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.0 GB | a621a4445242 |
| efficient_4xChinchilla_fp8/badam_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 94.8 KB | cd04d3afc692 |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.5 GB | 68abf3d46434 |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.5 GB | b48822c981ea |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.5 GB | fb16a6bc0ead |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.5 GB | f91373fbc804 |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.2 GB | 888affd714aa |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | d8f39ee0d627 |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.2 GB | 2c2a879b26e7 |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3b629813fc9b |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.2 GB | 2cdd94e08fde |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 5103d2045ed3 |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.2 GB | ef9d66ef0bf0 |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 9bb504619ab4 |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.5 GB | c15383ec9158 |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.5 GB | 34e71e4fb651 |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.5 GB | 731910b66b1a |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.5 GB | a9c3e2eb1311 |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/105975/main.pt | Weights | 2.7 GB | 2a22f04c494e |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/105975/worker_0.pt | Weights | 30.3 KB | 068900bea7cd |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/35325/main.pt | Weights | 2.7 GB | 099d386cd1d8 |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/35325/worker_0.pt | Weights | 228.5 KB | 3908eadfcbdb |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/70650/main.pt | Weights | 2.7 GB | 331c151bee6d |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/70650/worker_0.pt | Weights | 53.0 KB | 3ffb0fcde616 |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/latest/main.pt | Weights | 2.7 GB | 338a1c4e728b |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/ckpts/latest/worker_0.pt | Weights | 49.2 KB | 6553837aa147 |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_apollo_bfloat16_lr2e-3_wd0.1_b20.999_fineweb/ckpts/latest/main.pt | Weights | 2.2 GB | c14e13b9ade9 |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_apollo_bfloat16_lr2e-3_wd0.1_b20.999_fineweb/ckpts/latest/worker_0.pt | Weights | 18.1 KB | e99952d6034c |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_frugal_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/ckpts/latest/main.pt | Weights | 2.2 GB | 56d33d4f186d |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_frugal_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/ckpts/latest/worker_0.pt | Weights | 18.1 KB | e2739209c733 |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_galore_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/ckpts/latest/main.pt | Weights | 2.5 GB | aaa0f8756ceb |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_galore_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/ckpts/latest/worker_0.pt | Weights | 18.1 KB | e99952d6034c |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_slim_adam_bfloat16_lr5e-4_wd0.1_b20.999_fineweb/ckpts/latest/main.pt | Weights | 2.7 GB | e88a38c3b681 |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_slim_adam_bfloat16_lr5e-4_wd0.1_b20.999_fineweb/ckpts/latest/worker_0.pt | Weights | 64.6 KB | fdc5f464eab4 |
| llama500m_checkpoints/_corrupt/adamw_spectral_l1_cf1.4_lr1e-3_finewebedu/main.pt | Weights | 6.1 GB | 0382ea065550 |
| llama500m_checkpoints/_corrupt/adamw_spectral_l1_cf1.4_lr1e-3_finewebedu/worker_0.pt | Weights | 14.6 KB | bc0af491a5e6 |
| llama500m_checkpoints/adamw_lr1e-3_finewebedu/main.pt | Weights | 6.1 GB | ca805b86b112 |
| llama500m_checkpoints/adamw_lr1e-3_finewebedu/prehab_20260816/main.pt | Weights | 2.0 GB | 7361ab864bc3 |
| llama500m_checkpoints/adamw_lr1e-3_finewebedu/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/adamw_spectral_l1_cf0.7_lr1e-3_finewebedu/main.pt | Weights | 6.1 GB | 9726c0efe800 |
| llama500m_checkpoints/adamw_spectral_l1_cf0.7_lr1e-3_finewebedu/worker_0.pt | Weights | 14.6 KB | 777fae2b9841 |
| llama500m_checkpoints/adamw_spectral_l1_cf1.6_lr1e-3_finewebedu/main.pt | Weights | 6.1 GB | 938598f64c15 |
| llama500m_checkpoints/adamw_spectral_l1_cf1.6_lr1e-3_finewebedu/worker_0.pt | Weights | 14.6 KB | 109cb8bfa884 |
| llama500m_checkpoints/cuttlefish_rf0.5_fe100_lr1e-3_finewebedu/main.pt | Weights | 6.1 GB | cfb80b3eb0cc |
| llama500m_checkpoints/cuttlefish_rf0.5_fe100_lr1e-3_finewebedu/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.5_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | 55c822ea4744 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.5_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.5_finewebedu_erank/ckpts/latest/worker_1.pt | Weights | 14.7 KB | 8ca6e96e0a6d |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.7_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | f13b8ce29966 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.7_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | 57c85773c874 |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.2_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | b49328cdd5ab |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.2_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.4_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | ec960a7fb7b7 |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.4_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.6_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | 29ef1cfe3ede |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.6_finewebedu_erank/ckpts/latest/worker_0.pt | Weights | 14.6 KB | fa974c30d7a5 |
| spectral_l1_500m/llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.5_finewebedu_erank/ckpts/latest/main.pt | Weights | 6.1 GB | c7766758a3af |
| 1xChinchilla_fp8_resume/adamw_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| 1xChinchilla_fp8_resume/ademamix_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| 1xChinchilla_fp8_resume/muon_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| 1xChinchilla_fp8_resume/soap_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| 1xChinchilla_full_fp8_resume/500m_adamw_full_fp8_1xC_resume/summary.json | Configuration | 17.0 KB | — |
| 1xChinchilla_full_fp8_resume/500m_ademamix_full_fp8_1xC_resume/summary.json | Configuration | 17.1 KB | — |
| 1xChinchilla_full_fp8_resume/500m_muon_full_fp8_1xC_resume/summary.json | Configuration | 17.0 KB | — |
| 1xChinchilla_full_fp8_resume/500m_soap_full_fp8_1xC_resume/summary.json | Configuration | 17.0 KB | — |
| 1xChinchilla_optimizer_fp8/500m_adamw_optimizer_fp8_1xC/summary.json | Configuration | 85.2 KB | — |
| 1xChinchilla_optimizer_fp8/500m_soap_optimizer_fp8_1xC_bs16_acc8/summary.json | Configuration | 95.3 KB | — |
| 1xChinchilla_resume_1gpu/lion_muon_lr1e-3_1xC_resume_1gpu/summary.json | Configuration | 17.1 KB | — |
| 2xChinchilla_1gpu/lion_muon_lr1e-3_2xC_1gpu/summary.json | Configuration | 184.2 KB | — |
| 2xChinchilla_fp8_resume/adamw_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| 2xChinchilla_fp8_resume/ademamix_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| 2xChinchilla_fp8_resume/muon_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| 2xChinchilla_fp8_resume/soap_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| 2xChinchilla_full_fp8/500m_adamw_full_fp8_2xC/summary.json | Configuration | 184.2 KB | — |
| 2xChinchilla_full_fp8/500m_ademamix_full_fp8_2xC/summary.json | Configuration | 184.1 KB | — |
| 2xChinchilla_full_fp8/500m_apollo_full_fp8_2xC/summary.json | Configuration | 184.1 KB | — |
| 2xChinchilla_full_fp8/500m_frugal_full_fp8_2xC/summary.json | Configuration | 184.2 KB | — |
| 2xChinchilla_full_fp8/500m_galore_full_fp8_2xC/summary.json | Configuration | 184.3 KB | — |
| 2xChinchilla_full_fp8/500m_muon_full_fp8_2xC/summary.json | Configuration | 184.1 KB | — |
| 2xChinchilla_full_fp8/500m_soap_full_fp8_2xC/summary.json | Configuration | 184.1 KB | — |
| 4xChinchilla_fp8/adamw_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| 4xChinchilla_fp8/ademamix_act_fp8_4xC/summary.json | Configuration | 190.6 KB | — |
| 4xChinchilla_fp8/muon_act_fp8_4xC/summary.json | Configuration | 190.6 KB | — |
| 4xChinchilla_fp8/soap_act_fp8_4xC/summary.json | Configuration | 150.4 KB | — |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_bf16_wd1e-4_2xC_warmup2000_2gpu-iter-67911/SHA256.json | Configuration | 506 B | — |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_act_wd1e-4_2xC_warmup2000_2gpu-iter-67911/SHA256.json | Configuration | 509 B | — |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_fp8_full_wd1e-4_2xC_warmup2000_2gpu-iter-67911/SHA256.json | Configuration | 510 B | — |
| 500m-slimadam-precisions-20261002/llama500M_slim_adam_w16a16g32_fp8_states_wd1e-4_2xC_warmup2000_2gpu-iter-67911/SHA256.json | Configuration | 522 B | — |
| analysis/bootstrap_downstream_metrics_20260828/adamw_fp8_act/summary.json | Configuration | 755 B | — |
| analysis/bootstrap_downstream_metrics_20260828/adamw_fp8_states/summary.json | Configuration | 744 B | — |
| analysis/bootstrap_downstream_metrics_20260828/ademamix_fp8_act/summary.json | Configuration | 760 B | — |
| analysis/bootstrap_downstream_metrics_20260828/muon_fp8_act/summary.json | Configuration | 753 B | — |
| analysis/bootstrap_downstream_metrics_20260828/muon_fp8_states/summary.json | Configuration | 774 B | — |
| analysis/bootstrap_downstream_metrics_20260828/soap_fp8_states/summary.json | Configuration | 2.0 KB | — |
| analysis/bootstrap_downstream_metrics_20260828/soap_report/summary.json | Configuration | 748 B | — |
| analysis/bootstrap_downstream_metrics_20260828_invalid_wiki_batch128/soap_fp8_states/summary.json | Configuration | 2.0 KB | — |
| analysis/bootstrap_val_loss_ci/muon_2xc_seed2_itr77168/bootstrap_summary.json | Configuration | 2.1 KB | — |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/holdout_metadata.json | Configuration | 8.0 KB | — |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/paired_bootstrap_summary.json | Configuration | 11.9 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_act/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_states/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/ademamix_fp8_act/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_act/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_final_itr75457/bootstrap_summary.json | Configuration | 2.7 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_itr67911/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_itr67911/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/soap_fp8_states/bootstrap_summary.json | Configuration | 2.6 KB | — |
| analysis/hf_stage3_dense_predecay_upload.py | Configuration | 6.3 KB | — |
| analysis/hf_stage3_dense_rename_muon.py | Configuration | 5.1 KB | — |
| analysis/hf_stage3_dense_serial_upload.py | Configuration | 6.6 KB | — |
| analysis/hf_stage3_dense_upload.py | Configuration | 6.6 KB | — |
| analysis/hf_stage3_dense_upload/muon_fp8_act/metric_constraints.json | Configuration | 1.4 KB | — |
| analysis/report_rebootstrap_20260828/muon_fp8_act_itr77168/summary.json | Configuration | 2.1 KB | — |
| analysis/report_rebootstrap_20260828/soap_fp8_states_seed2/summary.json | Configuration | 2.0 KB | — |
| analysis/report_wikitext_ci_20260828/adamw_fp8_act/summary.json | Configuration | 1.0 KB | — |
| analysis/report_wikitext_ci_20260828/adamw_fp8_states/summary.json | Configuration | 1.0 KB | — |
| analysis/report_wikitext_ci_20260828/ademamix_fp8_act/summary.json | Configuration | 1.0 KB | — |
| analysis/report_wikitext_ci_20260828/muon_fp8_states/summary.json | Configuration | 1.0 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank/summary.json | Configuration | 16.3 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_5_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank/summary.json | Configuration | 16.2 KB | — |
| cf_bruteforce_124M/llama124m_adamw_wd0.1_lr5e-3_finewebedu_rankprofile_20261012/summary.json | Configuration | 4.9 KB | — |
| cf_bruteforce_124M/llama124m_adamw_wd0.5_lr5e-3_finewebedu_rankprofile_20261012/summary.json | Configuration | 4.9 KB | — |
| cf_bruteforce_124M/llama124m_adamw_wd1_lr5e-3_finewebedu_rankprofile_20261012/summary.json | Configuration | 4.9 KB | — |
| cf_bruteforce_124M/llama124m_adamw_wd2_lr5e-3_finewebedu_rankprofile_20261012/summary.json | Configuration | 4.9 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank/summary.json | Configuration | 28.0 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/GPU_SMOKE_OK.json | Configuration | 87 B | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank.json | Configuration | 1.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.1_lr5e-3_finewebedu_rankprofile_20261012.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.5_lr5e-3_finewebedu_rankprofile_20261012.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd1_lr5e-3_finewebedu_rankprofile_20261012.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd2_lr5e-3_finewebedu_rankprofile_20261012.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank.json | Configuration | 2.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank.json | Configuration | 2.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank.json | Configuration | 2.2 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank.json | Configuration | 2.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank.json | Configuration | 2.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank.json | Configuration | 2.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank.json | Configuration | 2.5 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank.json | Configuration | 2.3 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank.json | Configuration | 2.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank.json | Configuration | 2.4 MB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank.json | Configuration | 2.4 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/SUCCESS.json | Configuration | 75 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/llama124m_spectral_noembed_coef0.01.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/summary.json | Configuration | 4.8 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/SUCCESS.json | Configuration | 75 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/llama124m_spectral_noembed_coef0.1.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/summary.json | Configuration | 4.8 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.3_gpu_smoke/SUCCESS.json | Configuration | 71 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/SUCCESS.json | Configuration | 75 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/llama124m_spectral_noembed_coef0.json | Configuration | 1.3 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/summary.json | Configuration | 4.8 KB | — |
| efficient_1xChinchilla_fp8_resume/adam_mini_act_fp8_1xC_resume/summary.json | Configuration | 12.2 KB | — |
| efficient_1xChinchilla_fp8_resume/apollo_act_fp8_1xC_resume/summary.json | Configuration | 12.2 KB | — |
| efficient_1xChinchilla_fp8_resume/apollo_mini_act_fp8_1xC_resume/summary.json | Configuration | 12.2 KB | — |
| efficient_1xChinchilla_fp8_resume/fira_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| efficient_1xChinchilla_fp8_resume/frugal_act_fp8_1xC_resume/summary.json | Configuration | 12.1 KB | — |
| efficient_1xChinchilla_fp8_resume/galore_act_fp8_1xC_resume/summary.json | Configuration | 12.2 KB | — |
| efficient_1xChinchilla_fp8_resume/slim_adam_act_fp8_1xC_resume/summary.json | Configuration | 12.2 KB | — |
| efficient_2xChinchilla_fp8_resume/adam_mini_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| efficient_2xChinchilla_fp8_resume/apollo_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| efficient_2xChinchilla_fp8_resume/apollo_mini_act_fp8_2xC_resume/summary.json | Configuration | 16.8 KB | — |
| efficient_2xChinchilla_fp8_resume/fira_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| efficient_2xChinchilla_fp8_resume/frugal_act_fp8_2xC_resume/summary.json | Configuration | 16.8 KB | — |
| efficient_2xChinchilla_fp8_resume/galore_act_fp8_2xC_resume/summary.json | Configuration | 16.8 KB | — |
| efficient_2xChinchilla_fp8_resume/slim_adam_act_fp8_2xC_resume/summary.json | Configuration | 16.7 KB | — |
| efficient_4xChinchilla_fp8/adam_mini_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| efficient_4xChinchilla_fp8/apollo_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| efficient_4xChinchilla_fp8/apollo_mini_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| efficient_4xChinchilla_fp8/fira_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| efficient_4xChinchilla_fp8/frugal_act_fp8_4xC/summary.json | Configuration | 190.8 KB | — |
| efficient_4xChinchilla_fp8/galore_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| efficient_4xChinchilla_fp8/slim_adam_act_fp8_4xC/summary.json | Configuration | 190.7 KB | — |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_apollo_bfloat16_lr2e-3_wd0.1_b20.999_fineweb/summary.json | Configuration | 52.5 KB | — |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_frugal_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/summary.json | Configuration | 52.5 KB | — |
| llama-12L8H_fineweb_39k/2xBatch_llama257M_galore_bfloat16_lr1e-3_wd0.1_b20.999_fineweb/summary.json | Configuration | 52.5 KB | — |
| llama500m_checkpoints/_metadata/no_checkpoint/adamw_spectral_l1_cf0.5_lr1e-3_finewebedu/summary.json | Configuration | 28.3 KB | — |
| llama500m_checkpoints/_metadata/no_checkpoint/adamw_spectral_l1_cf1.4_lr1e-3_finewebedu/summary.json | Configuration | 28.3 KB | — |
| llama500m_checkpoints/adamw_lr1e-3_finewebedu/summary.json | Configuration | 16.0 KB | — |
| llama500m_checkpoints/adamw_spectral_l1_cf0.7_lr1e-3_finewebedu/summary.json | Configuration | 16.7 KB | — |
| llama500m_checkpoints/cuttlefish_rf0.5_fe100_lr1e-3_finewebedu/summary.json | Configuration | 16.0 KB | — |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.5_finewebedu_erank/summary.json | Configuration | 16.7 KB | — |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_0.7_finewebedu_erank/summary.json | Configuration | 16.5 KB | — |
| llama500m_checkpoints/h200_gpu2_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1_finewebedu_erank/summary.json | Configuration | 16.7 KB | — |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.2_finewebedu_erank/summary.json | Configuration | 16.5 KB | — |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.4_finewebedu_erank/summary.json | Configuration | 16.5 KB | — |
| llama500m_checkpoints/h200_gpu4_05ch_bs32acc4_llama500m_adamw-spectral-l1-reg_wd1e-1_lr1e-3_sl1_1.6_finewebedu_erank/summary.json | Configuration | 16.5 KB | — |
| analysis/hf_stage3_dense_upload/README.md | Documentation | 1.7 KB | — |
| analysis/__pycache__/hf_stage3_dense_predecay_upload.cpython-313.pyc | Other | 8.9 KB | — |
| analysis/__pycache__/hf_stage3_dense_rename_muon.cpython-313.pyc | Other | 6.9 KB | — |
| analysis/__pycache__/hf_stage3_dense_serial_upload.cpython-313.pyc | Other | 9.7 KB | — |
| analysis/__pycache__/hf_stage3_dense_upload.cpython-313.pyc | Other | 8.8 KB | — |
| analysis/bootstrap_downstream_metrics_20260828/soap_fp8_states/wikitext_per_sequence_losses.npy | Other | 2.0 KB | 9eaa48bc93ad |
| analysis/bootstrap_downstream_metrics_20260828_invalid_wiki_batch128/soap_fp8_states/wikitext_per_sequence_losses.npy | Other | 1.2 KB | 4569d6de33d7 |
| analysis/bootstrap_val_loss_ci/muon_2xc_seed2_itr77168/bootstrap_distribution.pdf | Other | 18.7 KB | — |
| analysis/bootstrap_val_loss_ci/muon_2xc_seed2_itr77168/bootstrap_distribution.png | Other | 55.0 KB | — |
| analysis/bootstrap_val_loss_ci/muon_2xc_seed2_itr77168/bootstrap_means.npy | Other | 800.1 KB | 12c71c96b5c3 |
| analysis/bootstrap_val_loss_ci/muon_2xc_seed2_itr77168/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/holdout_source/CC-MAIN-2025-26_000_00000.parquet | Other | 1.4 GB | 4cefc2e8d3ef |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/bf16_per_sequence_losses.npy | Other | 80.1 KB | 40675b525d99 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/fp8_per_sequence_losses.npy | Other | 80.1 KB | 730e4c524f3b |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/holdout_blocks.npy | Other | 41.0 MB | e256b6bea837 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/holdout_cluster_ids.npy | Other | 40.1 KB | 264a3a5c73f7 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/holdout_row_group_ids.npy | Other | 40.1 KB | 4749e97eb2b5 |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/paired_cluster_bootstrap.pdf | Other | 18.3 KB | — |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/paired_cluster_bootstrap.png | Other | 111.6 KB | f0b39a75cc6f |
| analysis/bootstrap_val_loss_ci/muon_fp8_77168_vs_bf16_75457/results/paired_per_sequence_losses.csv | Other | 627.6 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_act/bootstrap_distribution.pdf | Other | 17.9 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_act/bootstrap_distribution.png | Other | 49.7 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_act/bootstrap_means.npy | Other | 800.1 KB | 18257768f3ec |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_act/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_states/bootstrap_distribution.pdf | Other | 17.9 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_states/bootstrap_distribution.png | Other | 49.8 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_states/bootstrap_means.npy | Other | 800.1 KB | 26ff8fc2b1f6 |
| analysis/bootstrap_val_loss_fixed_reference_20260828/adamw_fp8_states/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/ademamix_fp8_act/bootstrap_distribution.pdf | Other | 18.1 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/ademamix_fp8_act/bootstrap_distribution.png | Other | 50.4 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/ademamix_fp8_act/bootstrap_means.npy | Other | 800.1 KB | 47333f81cbfe |
| analysis/bootstrap_val_loss_fixed_reference_20260828/ademamix_fp8_act/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_act/bootstrap_distribution.pdf | Other | 17.7 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_act/bootstrap_distribution.png | Other | 48.1 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_act/bootstrap_means.npy | Other | 800.1 KB | 8ad1341e1e62 |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_act/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states/bootstrap_distribution.pdf | Other | 17.8 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states/bootstrap_distribution.png | Other | 50.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states/bootstrap_means.npy | Other | 800.1 KB | 0215b08fad3a |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_final_itr75457/bootstrap_distribution.pdf | Other | 18.0 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_final_itr75457/bootstrap_distribution.png | Other | 50.1 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_final_itr75457/bootstrap_means.npy | Other | 800.1 KB | 055977e60d36 |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_final_itr75457/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_itr67911/bootstrap_distribution.pdf | Other | 17.8 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_itr67911/bootstrap_distribution.png | Other | 51.0 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_itr67911/bootstrap_means.npy | Other | 800.1 KB | 9c80c24b66b8 |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_cloud_itr67911/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_itr67911/bootstrap_distribution.pdf | Other | 17.8 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_itr67911/bootstrap_distribution.png | Other | 50.1 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_itr67911/bootstrap_means.npy | Other | 800.1 KB | 0ce41fece6ec |
| analysis/bootstrap_val_loss_fixed_reference_20260828/muon_fp8_states_itr67911/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/soap_fp8_states/bootstrap_distribution.pdf | Other | 18.5 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/soap_fp8_states/bootstrap_distribution.png | Other | 49.8 KB | — |
| analysis/bootstrap_val_loss_fixed_reference_20260828/soap_fp8_states/bootstrap_means.npy | Other | 800.1 KB | 83aefde85dc5 |
| analysis/bootstrap_val_loss_fixed_reference_20260828/soap_fp8_states/per_sequence_losses.csv | Other | 25.5 KB | — |
| analysis/hf_download_full_fp8_1xc.log | Other | 283 B | — |
| analysis/hf_stage3_dense_predecay_upload.log | Other | 151.8 MB | 3dfe59fc1dc9 |
| analysis/hf_stage3_dense_serial_upload.log | Other | 108.9 KB | — |
| analysis/hf_stage3_dense_upload.log | Other | 51.1 MB | 13e0e17e843d |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/README.md.metadata | Other | 143 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_act/main.pt.metadata | Other | 105 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_act/pre_decay/main.pt.metadata | Other | 106 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_act/pre_decay/worker_0.pt.metadata | Other | 100 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_act/worker_0.pt.metadata | Other | 102 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_states/main.pt.metadata | Other | 104 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_states/pre_decay/main.pt.metadata | Other | 105 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_states/pre_decay/worker_0.pt.metadata | Other | 101 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/adamw_fp8_states/worker_0.pt.metadata | Other | 102 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/ademamix_fp8_act/main.pt.metadata | Other | 104 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/ademamix_fp8_act/pre_decay/main.pt.metadata | Other | 106 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/ademamix_fp8_act/pre_decay/worker_0.pt.metadata | Other | 102 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/ademamix_fp8_act/worker_0.pt.metadata | Other | 100 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_act_latest_feasible/main.pt.metadata | Other | 106 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_act_latest_feasible/metric_constraints.json.metadata | Other | 104 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_act_latest_feasible/pre_decay/main.pt.metadata | Other | 105 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_act_latest_feasible/pre_decay/worker_0.pt.metadata | Other | 102 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_act_latest_feasible/worker_0.pt.metadata | Other | 102 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_states/main.pt.metadata | Other | 105 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_states/pre_decay/main.pt.metadata | Other | 106 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_states/pre_decay/worker_0.pt.metadata | Other | 101 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/muon_fp8_states/worker_0.pt.metadata | Other | 101 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/soap_fp8_states/main.pt.metadata | Other | 107 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/soap_fp8_states/pre_decay/main.pt.metadata | Other | 107 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/soap_fp8_states/pre_decay/worker_0.pt.metadata | Other | 101 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/upload/soap_fp8_states/worker_0.pt.metadata | Other | 101 B | — |
| analysis/report_rebootstrap_20260828/muon_fp8_act_itr77168/wikitext_per_sequence_losses.npy | Other | 1.9 KB | c40c91e32934 |
| analysis/report_rebootstrap_20260828/soap_fp8_states_seed2/wikitext_per_sequence_losses.npy | Other | 1.9 KB | 6872eb94bebf |
| analysis/report_wikitext_ci_20260828/adamw_fp8_act/wikitext_per_sequence_losses.npy | Other | 1.9 KB | 4c27021562a3 |
| analysis/report_wikitext_ci_20260828/adamw_fp8_states/wikitext_per_sequence_losses.npy | Other | 1.9 KB | 21cbfdc1332b |
| analysis/report_wikitext_ci_20260828/ademamix_fp8_act/wikitext_per_sequence_losses.npy | Other | 1.9 KB | 974323599767 |
| analysis/report_wikitext_ci_20260828/muon_fp8_states/wikitext_per_sequence_losses.npy | Other | 1.9 KB | 37a47323d79f |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank.csv | Other | 13.8 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.01_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank.csv | Other | 13.8 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.03_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank.csv | Other | 13.8 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.05_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank.csv | Other | 13.8 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.07_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.1_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.3_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.5_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0.7_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank.csv | Other | 13.5 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_0_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank.csv | Other | 13.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_10_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank.csv | Other | 13.6 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank.csv | Other | 13.6 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_2_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank.csv | Other | 13.6 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_3_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank.csv | Other | 13.6 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_4_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank.csv | Other | 13.6 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_7_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.1_lr5e-3_finewebedu_rankprofile_20261012.csv | Other | 12.8 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.1_lr5e-3_finewebedu_rankprofile_20261012_grouped.csv | Other | 1.0 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.5_lr5e-3_finewebedu_rankprofile_20261012.csv | Other | 12.9 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd0.5_lr5e-3_finewebedu_rankprofile_20261012_grouped.csv | Other | 1.0 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd1_lr5e-3_finewebedu_rankprofile_20261012.csv | Other | 12.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd1_lr5e-3_finewebedu_rankprofile_20261012_grouped.csv | Other | 1.0 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd2_lr5e-3_finewebedu_rankprofile_20261012.csv | Other | 12.7 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama124m_adamw_wd2_lr5e-3_finewebedu_rankprofile_20261012_grouped.csv | Other | 1.0 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.01_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank.csv | Other | 18.5 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.03_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank.csv | Other | 18.5 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.05_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank.csv | Other | 18.5 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.07_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank.csv | Other | 18.3 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.1_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank.csv | Other | 18.3 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.3_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank.csv | Other | 18.3 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.5_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0.7_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank.csv | Other | 18.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_0_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank.csv | Other | 18.2 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_1_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank.csv | Other | 18.3 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_2_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank.csv | Other | 18.2 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_3_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank.csv | Other | 18.3 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr3e-3_sl1_4_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.2_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.4_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.6_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank.csv | Other | 18.4 KB | — |
| cf_bruteforce_124M/rank_profiles_20261012/llama257m_adamw-spectral-l1-reg_wd1e-1_lr5e-3_sl1_1.8_finewebedu_erank_grouped.csv | Other | 1.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0.01/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0.1/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0.3/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0.5/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0.7/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef0/metrics.jsonl | Other | 237.4 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef1.3/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef1.6/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef1/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/attempt0/llama124m_spectral_noembed_coef2/metrics.jsonl | Other | — | |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/llama124m_spectral_noembed_coef0.01.csv | Other | 12.0 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/llama124m_spectral_noembed_coef0.01_grouped.csv | Other | 962 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.01/metrics.jsonl | Other | 2.7 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/llama124m_spectral_noembed_coef0.1.csv | Other | 11.9 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/llama124m_spectral_noembed_coef0.1_grouped.csv | Other | 955 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.1/metrics.jsonl | Other | 2.7 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.3/metrics.jsonl | Other | 1.0 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.5/metrics.jsonl | Other | 996.0 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0.7/metrics.jsonl | Other | 782.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/llama124m_spectral_noembed_coef0.csv | Other | 11.8 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/llama124m_spectral_noembed_coef0_grouped.csv | Other | 937 B | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef0/metrics.jsonl | Other | 2.7 MB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef1.3/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef1.6/metrics.jsonl | Other | 4.1 KB | — |
| cf_bruteforce_124M/spectral_noembed_20261012/llama124m_spectral_noembed_coef1/metrics.jsonl | Other | 4.1 KB | — |
| logs/500m_apollo_optimizer_fp8_1xC.log | Other | 7.2 MB | — |
| logs/500m_frugal_optimizer_fp8_1xC.log | Other | 7.2 MB | — |
| logs/500m_galore_optimizer_fp8_1xC.log | Other | 7.2 MB | — |
| logs/500m_slim_adam_optimizer_fp8_1xC.log | Other | 7.2 MB | — |
| .gitattributes | Repository | 1.8 KB | — |
| analysis/hf_stage3_dense_upload/.repo_created_by_uploader | Repository | 25 B | — |
| analysis/hf_stage3_dense_upload/_cache/huggingface/.gitignore | Repository | 1 B | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 699.4 GB
Download from Dmitrii Andriianov
Released by Dmitrii Andriianov through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 699.4 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.