archive e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975 (2026-10-20)
Browse files- .gitattributes +14 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/config.json +52 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/export_manifest.json +16 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/model.safetensors +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/.metadata +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__0_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__1_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__2_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__3_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__4_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__5_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/meta.json +81 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/.metadata +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__0_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__1_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__2_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__3_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__4_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__5_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/.metadata +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__0_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__1_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__2_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__3_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__4_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__5_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00000.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00001.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00002.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00003.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00004.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00005.pt +3 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00000.json +37 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00001.json +37 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00002.json +37 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00003.json +37 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00004.json +37 -0
- e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00005.json +37 -0
.gitattributes
CHANGED
|
@@ -831,3 +831,17 @@ e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/
|
|
| 831 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 832 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 833 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 831 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 832 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 833 |
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 834 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/.metadata filter=lfs diff=lfs merge=lfs -text
|
| 835 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 836 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 837 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 838 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 839 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 840 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 841 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/.metadata filter=lfs diff=lfs merge=lfs -text
|
| 842 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 843 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 844 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 845 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 846 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 847 |
+
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"DFlashDraftModel"
|
| 4 |
+
],
|
| 5 |
+
"attention_backend": "fa4",
|
| 6 |
+
"domain_anchored_token_moe": false,
|
| 7 |
+
"domain_order": null,
|
| 8 |
+
"draft_vocab_size": 151936,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_size": 4096,
|
| 11 |
+
"intermediate_size": 12288,
|
| 12 |
+
"learned_router": false,
|
| 13 |
+
"mask_token_id": 151669,
|
| 14 |
+
"max_position_embeddings": 40960,
|
| 15 |
+
"model_type": "dflash",
|
| 16 |
+
"num_attention_heads": 32,
|
| 17 |
+
"num_domains": 1,
|
| 18 |
+
"num_experts": 4,
|
| 19 |
+
"num_hidden_layers": 5,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"num_specialists": 4,
|
| 22 |
+
"num_target_layers": 5,
|
| 23 |
+
"per_domain_context_proj": false,
|
| 24 |
+
"per_domain_layer": false,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_theta": 1000000.0,
|
| 27 |
+
"router_init_std": 0.0015625,
|
| 28 |
+
"router_reads_shared": false,
|
| 29 |
+
"router_top_k": 1,
|
| 30 |
+
"specialist_dispatch": "grouped",
|
| 31 |
+
"specialist_full_width": true,
|
| 32 |
+
"specialist_gate_mode": "confidence",
|
| 33 |
+
"specialist_init_std": 0.02,
|
| 34 |
+
"specialist_rank": 512,
|
| 35 |
+
"specialist_top_k": 2,
|
| 36 |
+
"specialist_top_k_tau": 0.9,
|
| 37 |
+
"specialist_up_init_std": 0.001,
|
| 38 |
+
"target_hidden_size": 4096,
|
| 39 |
+
"target_layer_ids": [
|
| 40 |
+
1,
|
| 41 |
+
9,
|
| 42 |
+
17,
|
| 43 |
+
25,
|
| 44 |
+
33
|
| 45 |
+
],
|
| 46 |
+
"target_num_hidden_layers": 36,
|
| 47 |
+
"tie_word_embeddings": false,
|
| 48 |
+
"token_moe": true,
|
| 49 |
+
"token_moe_shared": true,
|
| 50 |
+
"transformers_version": "4.57.1",
|
| 51 |
+
"vocab_size": 151936
|
| 52 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/export_manifest.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"distributed_checkpoint_is_resume_authority": true,
|
| 3 |
+
"domain_anchored_token_moe": false,
|
| 4 |
+
"format": "dflash-hf-sglang",
|
| 5 |
+
"num_experts": 4,
|
| 6 |
+
"num_hidden_layers": 5,
|
| 7 |
+
"request_router_checkpoint_sha256": null,
|
| 8 |
+
"request_router_manifest_sha256": null,
|
| 9 |
+
"requires_external_request_router": false,
|
| 10 |
+
"router_semantics": "per-token-per-layer-hard-top1",
|
| 11 |
+
"router_top_k": 1,
|
| 12 |
+
"tensor_count": 128,
|
| 13 |
+
"tensor_schema_sha256": "sha256:f6a560a19d633906c790f87a9a4d76374266577061b2eba8052bddfd8b1178f6",
|
| 14 |
+
"token_moe": true,
|
| 15 |
+
"trainer_global_step": 39975
|
| 16 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/hf/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2ec2f1306921f9a4cb997f43fddba6c4616600beb658d4fad9e59a8a2abaa71
|
| 3 |
+
size 8137392760
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/.metadata
ADDED
|
Binary file (3.45 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__0_0.distcp
ADDED
|
Binary file (3.78 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__1_0.distcp
ADDED
|
Binary file (3.78 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__2_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__3_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__4_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/lr_scheduler/__5_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/meta.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval/acc_0": 0.8162078419134878,
|
| 3 |
+
"eval/acc_1": 0.6796893323345857,
|
| 4 |
+
"eval/acc_10": 0.1999944092451673,
|
| 5 |
+
"eval/acc_11": 0.18266130210598913,
|
| 6 |
+
"eval/acc_12": 0.16721169962441124,
|
| 7 |
+
"eval/acc_13": 0.15306371448321698,
|
| 8 |
+
"eval/acc_14": 0.14180343980005888,
|
| 9 |
+
"eval/acc_2": 0.5734755490243739,
|
| 10 |
+
"eval/acc_3": 0.4877325767829167,
|
| 11 |
+
"eval/acc_4": 0.41917440807273504,
|
| 12 |
+
"eval/acc_5": 0.362737992231453,
|
| 13 |
+
"eval/acc_6": 0.3160810297208069,
|
| 14 |
+
"eval/acc_7": 0.27740628730340927,
|
| 15 |
+
"eval/acc_8": 0.24682935848798543,
|
| 16 |
+
"eval/acc_9": 0.22037984768540758,
|
| 17 |
+
"eval/accept_length": 4.535188496337764,
|
| 18 |
+
"eval/accepted_draft_tokens": 3.5351884963377644,
|
| 19 |
+
"eval/avg_acc": 0.3500578040575119,
|
| 20 |
+
"eval/avg_loss": 2.371686319900615,
|
| 21 |
+
"eval/block_count": 251730,
|
| 22 |
+
"eval/domain/code/accept_length": 4.909482170746919,
|
| 23 |
+
"eval/domain/code/accepted_draft_tokens": 3.90948217074692,
|
| 24 |
+
"eval/domain/code/avg_acc": 0.37755335321045635,
|
| 25 |
+
"eval/domain/code/avg_loss": 2.0681075324859317,
|
| 26 |
+
"eval/domain/code/block_count": 51117,
|
| 27 |
+
"eval/domain/code/prompt_count": 100,
|
| 28 |
+
"eval/domain/creative_writing/accept_length": 3.3859400523508687,
|
| 29 |
+
"eval/domain/creative_writing/accepted_draft_tokens": 2.3859400523508687,
|
| 30 |
+
"eval/domain/creative_writing/avg_acc": 0.26939246190968713,
|
| 31 |
+
"eval/domain/creative_writing/avg_loss": 3.024274484633321,
|
| 32 |
+
"eval/domain/creative_writing/block_count": 50811,
|
| 33 |
+
"eval/domain/creative_writing/prompt_count": 100,
|
| 34 |
+
"eval/domain/factual_qa/accept_length": 3.6248991459786017,
|
| 35 |
+
"eval/domain/factual_qa/accepted_draft_tokens": 2.6248991459786017,
|
| 36 |
+
"eval/domain/factual_qa/avg_acc": 0.28727952443875693,
|
| 37 |
+
"eval/domain/factual_qa/avg_loss": 2.796578022178046,
|
| 38 |
+
"eval/domain/factual_qa/block_count": 49576,
|
| 39 |
+
"eval/domain/factual_qa/prompt_count": 100,
|
| 40 |
+
"eval/domain/general/accept_length": 4.276281175438955,
|
| 41 |
+
"eval/domain/general/accepted_draft_tokens": 3.2762811754389545,
|
| 42 |
+
"eval/domain/general/avg_acc": 0.332619687829027,
|
| 43 |
+
"eval/domain/general/avg_loss": 2.4733561279146428,
|
| 44 |
+
"eval/domain/general/block_count": 49037,
|
| 45 |
+
"eval/domain/general/prompt_count": 100,
|
| 46 |
+
"eval/domain/math/accept_length": 6.431811522006681,
|
| 47 |
+
"eval/domain/math/accepted_draft_tokens": 5.431811522006681,
|
| 48 |
+
"eval/domain/math/avg_acc": 0.47948926288215504,
|
| 49 |
+
"eval/domain/math/avg_loss": 1.5214698816450631,
|
| 50 |
+
"eval/domain/math/block_count": 51189,
|
| 51 |
+
"eval/domain/math/prompt_count": 100,
|
| 52 |
+
"eval/macro5/accept_length": 4.525682813304405,
|
| 53 |
+
"eval/macro5/accepted_draft_tokens": 3.5256828133044054,
|
| 54 |
+
"eval/macro5/avg_acc": 0.34926685805401647,
|
| 55 |
+
"eval/macro5/avg_loss": 2.3767572097714007,
|
| 56 |
+
"eval/ploss_0": 0.5795250160334666,
|
| 57 |
+
"eval/ploss_1": 1.1708491212384111,
|
| 58 |
+
"eval/ploss_10": 3.9068404996227337,
|
| 59 |
+
"eval/ploss_11": 4.033615396795811,
|
| 60 |
+
"eval/ploss_12": 4.147099807119658,
|
| 61 |
+
"eval/ploss_13": 4.249743694314163,
|
| 62 |
+
"eval/ploss_14": 4.341899055645657,
|
| 63 |
+
"eval/ploss_2": 1.711860161367696,
|
| 64 |
+
"eval/ploss_3": 2.1779804254563295,
|
| 65 |
+
"eval/ploss_4": 2.574671651869955,
|
| 66 |
+
"eval/ploss_5": 2.902747588991017,
|
| 67 |
+
"eval/ploss_6": 3.1786747374527784,
|
| 68 |
+
"eval/ploss_7": 3.411725197115745,
|
| 69 |
+
"eval/ploss_8": 3.6013609675398426,
|
| 70 |
+
"eval/ploss_9": 3.768708985909885,
|
| 71 |
+
"eval/prompt_count": 500,
|
| 72 |
+
"eval/wallclock_seconds": 20.251509526744485,
|
| 73 |
+
"global_step": 39975,
|
| 74 |
+
"inference_id": 39975,
|
| 75 |
+
"iteration": 39975,
|
| 76 |
+
"next_inference_id": 39975,
|
| 77 |
+
"next_step": 39975,
|
| 78 |
+
"step": 39975,
|
| 79 |
+
"timestamp": 1792512422.9219983,
|
| 80 |
+
"world_size": 6
|
| 81 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/.metadata
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0a1dcb959c389c803aae257460c89bf8786631ea2b3dec635f34a636078124a
|
| 3 |
+
size 179737
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7884eb54d3bd51f11a8d88fab837611e7b5b0c5687cfd38684d842a9a279d37e
|
| 3 |
+
size 1564178857
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:feb61a3068e321d5ffff910db40f819e7d53c101bc1de9b5023c0a79c8111f1b
|
| 3 |
+
size 1564178857
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42cd7fd149a257cf7ba70ace4ef897d27931d311d1f1fbbb1af4b2237644dd30
|
| 3 |
+
size 1564178857
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6853a2827f6910e38858214c7270ebdac77a980534396088c0f2690622230e8
|
| 3 |
+
size 1564178857
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2e65695651c33938d840f8f12cc5efce5690bc6a63e79071cc4f7b2a8b2c7e3
|
| 3 |
+
size 1564081167
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/model/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5af0e08401d86143701aa9c48b59e5ce0eea737f2d17e8e751c625f3331c39d
|
| 3 |
+
size 1562441513
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/.metadata
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff4ef1210dc6390b103bc6f7ae37aa1b5d3e753feb648d8d0e91b47dbd8c1092
|
| 3 |
+
size 468667
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb74762d98d8fb1cad28c63c860042c00bd3878498de8eaca1bd5ac5eb0ca259
|
| 3 |
+
size 8139539753
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c60b1d44e1e541e509220aa37e26f01462f4159185d659a835cdef47ea0c492
|
| 3 |
+
size 8139590655
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d895d15c851477d365f4acc0467ea62689353517f746af1c7cf4270c9e10438
|
| 3 |
+
size 8139589266
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de93620f08c3be28838536bb7e5930cf65045f6dd515b732ea67c62e17f37558
|
| 3 |
+
size 8139592733
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9075a19544a97abfb13bf5de5cf5bd81a3fd3b63f0ca41896e5d2b0bcd55e0a6
|
| 3 |
+
size 8139299663
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/optimizer/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d148aa649d23040c4582382c40fb441259f01327a5aeec588703c76e3d3904aa
|
| 3 |
+
size 8129514781
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffdc0062f2aa32f6ab8de1c4f6a4fadbcb4be5ec036f5d5546ce4457a369a436
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00001.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b055932a7302b8de10f0e2589924fe3cf553dfad9e57b628b7a2d73bc570fe4a
|
| 3 |
+
size 14773
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00002.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6d11d4d8aa09dfb7966af7340f7e26c3bbddc34f69e03793216a515abbfa282
|
| 3 |
+
size 14773
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00003.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9891b82fadab7bfbcb8d28e40f5498063536fa79b49dbd6db087e0368fcd4b45
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00004.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7624d4a1c249f3318e2e223aa7091f5a0ca21cba11a764bf203eea835500b189
|
| 3 |
+
size 14773
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/rng_rank_00005.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3bd28416e4065c616d388801c002f7df1d26cf85f97b51f653dd30cea497609
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00000.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "785256a337472570ac20d2f8ea55f10dd8436b954aedb77df3676483da7a06d0",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "823977aa79819c287e69136e25121145e40398eb1ae458ab7532351165b0961d",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "a5169229897a470b5f9edff12d3c2189ecfce2c4f9baea4c3df5a3720fe04a3b",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 0,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00001.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "34bfc0d009bcb4dff1d2595cdd601c1cc4c3c8bce488472b7a0b7940106c05b7",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "59ea21104746d87c848fb96241c012b2357bd8f7ee1300a4365d3dcccc98c43a",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "58884c2f27e97b0db6f5045bc39a4fdec166cc83c1a76d874fdb00a4327f0e48",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 1,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00002.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "abd26a5d1fc4395c4d50afcf329fdd692d673fa911d992495422fc6a7186ee73",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "f1bbd6ba22d09ae42ea4850753c75ccd163ea4e6a859c6763872de2069c514a6",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "b3be9e62e893148272bdec4ff7e55d36d7ed6ccfcae890662a4e11eb7bfaead3",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 2,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00003.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "21d3155659c153aa07981bc2d0886c7bad689cd5b1579a17883326b43c6965c8",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "dc2f7994daf3c7159346565c91c967cbc3e70db16f6d8677443ccd60fbd0b848",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "129fc8c9010890dec8fcd1213c16976ab8af97b74dd6c10fe60f264a2defb79a",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 3,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00004.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "ae03bed93d05953268130ebf1e357a382b1a2d1a229b78d172b73d80c9bb839d",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "919b259b18ccea9742ded0b9ffdfcd7a5abc23502f257a1474873fe9e54df851",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "7abb3de09e73e8cfd17bf024c45e134f1bd9839d1e410e1d2e030485c13ae88b",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 4,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0039975/state_fingerprint_rank_00005.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"components": {
|
| 3 |
+
"data": {
|
| 4 |
+
"sampled_tensor_count": 0,
|
| 5 |
+
"sha256": "322594c96bcee57c77a5b7a138a9e672d5f3a23b976ba63295b5e030e84e4295",
|
| 6 |
+
"tensor_count": 0
|
| 7 |
+
},
|
| 8 |
+
"model": {
|
| 9 |
+
"sampled_tensor_count": 107,
|
| 10 |
+
"sha256": "5116e9c3e9381980b1036159cda2e5b2ec5f7f53b81690ec71b6a779efab3046",
|
| 11 |
+
"tensor_count": 144
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 288,
|
| 15 |
+
"sha256": "2109f9eb6acd146c46deee34d09cef1895e284e565a280d339c11badf66c22ad",
|
| 16 |
+
"tensor_count": 512
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "f08fb76235fe07f97fa8c975f83bded741d13f8f1e25ecdd96cad1acedf384b6",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "e6474f432c20c056fba0655620e86830e90a995168fc07ca04a4eca8ed4da2a5",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 39975,
|
| 30 |
+
"rank": 5,
|
| 31 |
+
"sampling": {
|
| 32 |
+
"online_batch_max_elements_per_tensor": null,
|
| 33 |
+
"state_max_elements_per_tensor": 8192
|
| 34 |
+
},
|
| 35 |
+
"schema_version": 1,
|
| 36 |
+
"world_size": 6
|
| 37 |
+
}
|