archive e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325 (2026-10-20)
Browse files- .gitattributes +13 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/hf/config.json +47 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/hf/export_manifest.json +16 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/hf/model.safetensors +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/.metadata +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__0_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__1_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__2_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__3_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__4_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__5_0.distcp +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/meta.json +81 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/.metadata +0 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__0_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__1_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__2_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__3_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__4_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__5_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/.metadata +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__0_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__1_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__2_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__3_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__4_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__5_0.distcp +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00000.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00001.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00002.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00003.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00004.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00005.pt +3 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00000.json +37 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00001.json +37 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00002.json +37 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00003.json +37 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00004.json +37 -0
- e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/state_fingerprint_rank_00005.json +37 -0
.gitattributes
CHANGED
|
@@ -818,3 +818,16 @@ e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_001
|
|
| 818 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 819 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 820 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 818 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 819 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 820 |
e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_0013325/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 821 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 822 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 823 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 824 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 825 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 826 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 827 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/.metadata filter=lfs diff=lfs merge=lfs -text
|
| 828 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 829 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 830 |
+
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__2_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
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/hf/config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1,
|
| 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": 0.02,
|
| 28 |
+
"router_reads_shared": false,
|
| 29 |
+
"router_top_k": 1,
|
| 30 |
+
"specialist_init_std": 0.02,
|
| 31 |
+
"specialist_rank": 512,
|
| 32 |
+
"specialist_up_init_std": 0.0,
|
| 33 |
+
"target_hidden_size": 4096,
|
| 34 |
+
"target_layer_ids": [
|
| 35 |
+
1,
|
| 36 |
+
9,
|
| 37 |
+
17,
|
| 38 |
+
25,
|
| 39 |
+
33
|
| 40 |
+
],
|
| 41 |
+
"target_num_hidden_layers": 36,
|
| 42 |
+
"tie_word_embeddings": false,
|
| 43 |
+
"token_moe": false,
|
| 44 |
+
"token_moe_shared": false,
|
| 45 |
+
"transformers_version": "4.57.1",
|
| 46 |
+
"vocab_size": 151936
|
| 47 |
+
}
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 1,
|
| 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": "dense-shared-mlp",
|
| 11 |
+
"router_top_k": 1,
|
| 12 |
+
"tensor_count": 58,
|
| 13 |
+
"tensor_schema_sha256": "sha256:067ce575542e1d222a52c64c25b7fe53cc6d0f3c77928e6941d78057ff2e5a5f",
|
| 14 |
+
"token_moe": false,
|
| 15 |
+
"trainer_global_step": 13325
|
| 16 |
+
}
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/hf/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b8e8a3121529018f316db70a002b522a9014999901bfdaeeb55f2956e65147d
|
| 3 |
+
size 2097259072
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/.metadata
ADDED
|
Binary file (3.45 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__0_0.distcp
ADDED
|
Binary file (3.78 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__1_0.distcp
ADDED
|
Binary file (3.78 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__2_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__3_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__4_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/lr_scheduler/__5_0.distcp
ADDED
|
Binary file (2.52 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/meta.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval/acc_0": 0.8183410791415393,
|
| 3 |
+
"eval/acc_1": 0.6842364678766931,
|
| 4 |
+
"eval/acc_10": 0.20330921180754336,
|
| 5 |
+
"eval/acc_11": 0.18543440704563122,
|
| 6 |
+
"eval/acc_12": 0.16967434012509747,
|
| 7 |
+
"eval/acc_13": 0.15466766958353084,
|
| 8 |
+
"eval/acc_14": 0.14372846399528946,
|
| 9 |
+
"eval/acc_2": 0.5766570835956099,
|
| 10 |
+
"eval/acc_3": 0.4924437772737532,
|
| 11 |
+
"eval/acc_4": 0.4229842881373547,
|
| 12 |
+
"eval/acc_5": 0.3672628070183658,
|
| 13 |
+
"eval/acc_6": 0.31963989072703236,
|
| 14 |
+
"eval/acc_7": 0.2812620874602497,
|
| 15 |
+
"eval/acc_8": 0.2509066110778186,
|
| 16 |
+
"eval/acc_9": 0.22392444598483105,
|
| 17 |
+
"eval/accept_length": 4.58687879895537,
|
| 18 |
+
"eval/accepted_draft_tokens": 3.5868787989553694,
|
| 19 |
+
"eval/avg_acc": 0.3533938904623952,
|
| 20 |
+
"eval/avg_loss": 2.3479046224610154,
|
| 21 |
+
"eval/block_count": 251730,
|
| 22 |
+
"eval/domain/code/accept_length": 4.956902791634876,
|
| 23 |
+
"eval/domain/code/accepted_draft_tokens": 3.9569027916348767,
|
| 24 |
+
"eval/domain/code/avg_acc": 0.3796196765500562,
|
| 25 |
+
"eval/domain/code/avg_loss": 2.0547197602376674,
|
| 26 |
+
"eval/domain/code/block_count": 51117,
|
| 27 |
+
"eval/domain/code/prompt_count": 100,
|
| 28 |
+
"eval/domain/creative_writing/accept_length": 3.43689358603452,
|
| 29 |
+
"eval/domain/creative_writing/accepted_draft_tokens": 2.43689358603452,
|
| 30 |
+
"eval/domain/creative_writing/avg_acc": 0.2737982656657257,
|
| 31 |
+
"eval/domain/creative_writing/avg_loss": 2.9958842683144433,
|
| 32 |
+
"eval/domain/creative_writing/block_count": 50811,
|
| 33 |
+
"eval/domain/creative_writing/prompt_count": 100,
|
| 34 |
+
"eval/domain/factual_qa/accept_length": 3.6801678243713725,
|
| 35 |
+
"eval/domain/factual_qa/accepted_draft_tokens": 2.6801678243713725,
|
| 36 |
+
"eval/domain/factual_qa/avg_acc": 0.29118527211947487,
|
| 37 |
+
"eval/domain/factual_qa/avg_loss": 2.7602163231943555,
|
| 38 |
+
"eval/domain/factual_qa/block_count": 49576,
|
| 39 |
+
"eval/domain/factual_qa/prompt_count": 100,
|
| 40 |
+
"eval/domain/general/accept_length": 4.324632420417236,
|
| 41 |
+
"eval/domain/general/accepted_draft_tokens": 3.324632420417236,
|
| 42 |
+
"eval/domain/general/avg_acc": 0.3366680830533231,
|
| 43 |
+
"eval/domain/general/avg_loss": 2.4417283385511235,
|
| 44 |
+
"eval/domain/general/block_count": 49037,
|
| 45 |
+
"eval/domain/general/prompt_count": 100,
|
| 46 |
+
"eval/domain/math/accept_length": 6.488229893141105,
|
| 47 |
+
"eval/domain/math/accepted_draft_tokens": 5.488229893141105,
|
| 48 |
+
"eval/domain/math/avg_acc": 0.4818064109330754,
|
| 49 |
+
"eval/domain/math/avg_loss": 1.5115162304246699,
|
| 50 |
+
"eval/domain/math/block_count": 51189,
|
| 51 |
+
"eval/domain/math/prompt_count": 100,
|
| 52 |
+
"eval/macro5/accept_length": 4.577365303119822,
|
| 53 |
+
"eval/macro5/accepted_draft_tokens": 3.577365303119822,
|
| 54 |
+
"eval/macro5/avg_acc": 0.352615541664331,
|
| 55 |
+
"eval/macro5/avg_loss": 2.352812984144452,
|
| 56 |
+
"eval/ploss_0": 0.571868653832642,
|
| 57 |
+
"eval/ploss_1": 1.1575154668405763,
|
| 58 |
+
"eval/ploss_10": 3.875247761986614,
|
| 59 |
+
"eval/ploss_11": 4.001824694439553,
|
| 60 |
+
"eval/ploss_12": 4.118399855402735,
|
| 61 |
+
"eval/ploss_13": 4.220410376273289,
|
| 62 |
+
"eval/ploss_14": 4.316318908378732,
|
| 63 |
+
"eval/ploss_2": 1.6943838422468371,
|
| 64 |
+
"eval/ploss_3": 2.1521525271653723,
|
| 65 |
+
"eval/ploss_4": 2.546431227145111,
|
| 66 |
+
"eval/ploss_5": 2.867928959171316,
|
| 67 |
+
"eval/ploss_6": 3.143852324977737,
|
| 68 |
+
"eval/ploss_7": 3.377758036692556,
|
| 69 |
+
"eval/ploss_8": 3.5664602808082986,
|
| 70 |
+
"eval/ploss_9": 3.7331303567268366,
|
| 71 |
+
"eval/prompt_count": 500,
|
| 72 |
+
"eval/wallclock_seconds": 13.167444042861462,
|
| 73 |
+
"global_step": 13325,
|
| 74 |
+
"inference_id": 13325,
|
| 75 |
+
"iteration": 13325,
|
| 76 |
+
"next_inference_id": 13325,
|
| 77 |
+
"next_step": 13325,
|
| 78 |
+
"step": 13325,
|
| 79 |
+
"timestamp": 1792271192.9264855,
|
| 80 |
+
"world_size": 6
|
| 81 |
+
}
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/.metadata
ADDED
|
Binary file (80.4 kB). View file
|
|
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc67edad15e15a794f97e2d646ab58508155349dc0690564ec2c7d8ca25a46e3
|
| 3 |
+
size 557189747
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd355605376b123d240036fef3170673a3baa65b68d670a8073f8162b5cc911b
|
| 3 |
+
size 557189747
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a82a074687e1ad01646f135b4122c022d2b4e1bc7033e6d989863e7f2d4a8e5
|
| 3 |
+
size 557189747
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf4f110367c67a19a079c3b1a5f9ebff1b0a3554710249b711ab07d5dcf8f33e
|
| 3 |
+
size 557189747
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70280399ac2a931034627036de3ddde20bb5cdc854deb7114d1af1edefe1286c
|
| 3 |
+
size 557189747
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/model/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16ce2270c392d5b470b155c081f00e74423759742c5ddab11a74e335ca1cca48
|
| 3 |
+
size 556518003
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/.metadata
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8162b972da058711eaa21d66491389fd5611a9be1683714eb5f80b1585310d2c
|
| 3 |
+
size 208981
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f3bd5a0d58dd7faaf5e6d9d98ef9493e77851022f495ebccc9a5dc6460cf2ec
|
| 3 |
+
size 2098182023
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c5721d10fe8f85af5c9219b5a0d3f9f9f2933dd481ca8eb3d4e62e58bfcccbd
|
| 3 |
+
size 2098200759
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:307ee0397e6dd40ed8a224d2a3999169d2f790a375c0bcc2425919133d6f973b
|
| 3 |
+
size 2098199370
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:606fa426343824c6e3bd6ad28732ecfb57555ef18dcc324381b6d4d158b1040d
|
| 3 |
+
size 2098202837
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79d0c3771db64c2dad9e77750f87ae138880e00c025ff148d3f8236bef998245
|
| 3 |
+
size 2098202837
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/optimizer/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bfcc1c3a3617e434aa9421355d9a69a9291649dd9cc313b057104a85ce3c2db
|
| 3 |
+
size 2094270677
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54011eaaaf00be5a62d210ce0434a75fbb2b5502449e10eb1556b8b554087463
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00001.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dfda430f82aa3a6d6c028b8d707fb6dccf1310d094bdc058485a70fd96b05b4
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00002.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c787726f06748821b77cd8fe934346f4cc208cbdb7df2f1626b0b933157f728
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00003.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e3974b808ff6cf3476b440ed24c1302fbb269655f847b71e7a05bfbf0240be1
|
| 3 |
+
size 14773
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00004.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0aa80bdb62a0c9e68b74502fabd02a52d03890ee957b9f3422181d72f1e577cf
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/rng_rank_00005.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9c76b87d77942466cf25bc30cc9ade55b55dca45075c9cec5d77b27f2780d8f
|
| 3 |
+
size 14709
|
e5_8b_800k_20260831/arms/e5-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "25fc6fd5cccc1fe1093d889e6f0aac946c4727310ca1143668bbbe743eb80efb",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "8f77aa7ef6c3e647501874b437e5d5961cd7f5c437a07d3452f3149d9fb0dbe2",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "ccf4f94a4a98676cb95988b4412f9d65bf01f70f2a83fc8b8f5d60e2606bf564",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "edf08966d546169fca83c7a4aa0b35c3aa597bc337ee84d529ad1df1245f394c",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "c688268c682660c0ba3add9a99f62549749701dde77e9ad5b6e3fc05edb79c19",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "00de449c36739c62e45ca011ab0536070974e843b77c554e9a7bb801ff0739a4",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "5d7df8d49c56d23bd4ac9ec6a520fb4bded6e5dd71264924656da1a7624495cd",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "b3a991c17ba8f5112ecb8fd57cd1d5883fbf445545e43077d4a4351b04400da2",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "c27e22127ba4d2a73b94acfb619dbaf0eaab111b3554f2a32442dd66a361fea5",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "fcd33257948f7c6c7699ef827318f63fa3e5060bb7bfc48c6d107b73225e9f88",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "c4362958734ddbf62ebaf3570206454a0e2e27c7670d87bce138eca15feeaf37",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "336703d116460c1691825e86ed0871991c19c2a002396094629c22c5a499552d",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "bbd7749f6308f88295380cc4a9fd7735ba66aa419f00070ab39cc3695741a131",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "29e13e946c2c17668ce788c27e6d746099fbebf70bccd2c950e8deaa5a00951e",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "a21f0d9ff82f2f708a21b8524b3c6de5910757068be4002a66d5e5f27670dc16",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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-dense-lr1e3-cont1ep-anneal/checkpoints/iter_0013325/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": 47,
|
| 10 |
+
"sha256": "9a6224b2ac11546b66ff142bb723cedaf695e237f3afb2fce28247e9318865e5",
|
| 11 |
+
"tensor_count": 74
|
| 12 |
+
},
|
| 13 |
+
"optimizer": {
|
| 14 |
+
"sampled_tensor_count": 108,
|
| 15 |
+
"sha256": "065f3198d4527a5bd28ae8d5e3d7ae1c87567ea2cc4758cb8f1c25c2b41dcdac",
|
| 16 |
+
"tensor_count": 232
|
| 17 |
+
},
|
| 18 |
+
"rng": {
|
| 19 |
+
"sampled_tensor_count": 0,
|
| 20 |
+
"sha256": "73751e24e0659581241fdd415f65bcc78913512236dbe4386928a914db6f77d1",
|
| 21 |
+
"tensor_count": 2
|
| 22 |
+
},
|
| 23 |
+
"scheduler": {
|
| 24 |
+
"sampled_tensor_count": 0,
|
| 25 |
+
"sha256": "523af7de804690abffbc1ca8265c8df4a820b6d391e90e8b327f4af3ab522f23",
|
| 26 |
+
"tensor_count": 0
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"global_step": 13325,
|
| 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 |
+
}
|