ryan-0608 commited on
Commit
e8fd2b3
·
verified ·
1 Parent(s): f66ee2e

archive e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650 (2026-10-21)

Browse files
Files changed (38) hide show
  1. .gitattributes +13 -0
  2. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/hf/config.json +47 -0
  3. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/hf/export_manifest.json +16 -0
  4. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/hf/model.safetensors +3 -0
  5. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/.metadata +0 -0
  6. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__0_0.distcp +0 -0
  7. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__1_0.distcp +0 -0
  8. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__2_0.distcp +0 -0
  9. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__3_0.distcp +0 -0
  10. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__4_0.distcp +0 -0
  11. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__5_0.distcp +0 -0
  12. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/meta.json +81 -0
  13. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/.metadata +0 -0
  14. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__0_0.distcp +3 -0
  15. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__1_0.distcp +3 -0
  16. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__2_0.distcp +3 -0
  17. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__3_0.distcp +3 -0
  18. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__4_0.distcp +3 -0
  19. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__5_0.distcp +3 -0
  20. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/.metadata +3 -0
  21. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__0_0.distcp +3 -0
  22. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__1_0.distcp +3 -0
  23. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__2_0.distcp +3 -0
  24. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__3_0.distcp +3 -0
  25. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__4_0.distcp +3 -0
  26. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__5_0.distcp +3 -0
  27. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00000.pt +3 -0
  28. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00001.pt +3 -0
  29. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00002.pt +3 -0
  30. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00003.pt +3 -0
  31. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00004.pt +3 -0
  32. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00005.pt +3 -0
  33. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00000.json +37 -0
  34. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00001.json +37 -0
  35. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00002.json +37 -0
  36. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00003.json +37 -0
  37. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00004.json +37 -0
  38. e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/state_fingerprint_rank_00005.json +37 -0
.gitattributes CHANGED
@@ -845,3 +845,16 @@ e5_8b_800k_20260831/arms/e5-x4rand-sharedlr1e3-explr2e3-mos/checkpoints/iter_003
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
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
848
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
849
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
850
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
851
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
852
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
853
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
854
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/.metadata filter=lfs diff=lfs merge=lfs -text
855
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
856
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
857
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
858
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
859
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
860
+ e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/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-lr2e3/checkpoints/iter_0026650/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": 26650
16
+ }
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/hf/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fb5e9e3738bc745396f9d4749dd857eba76858434367f09de086c82cc0ade89
3
+ size 2097259072
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/.metadata ADDED
Binary file (3.43 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__0_0.distcp ADDED
Binary file (3.78 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__1_0.distcp ADDED
Binary file (3.78 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__2_0.distcp ADDED
Binary file (2.52 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__3_0.distcp ADDED
Binary file (2.52 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__4_0.distcp ADDED
Binary file (2.52 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/lr_scheduler/__5_0.distcp ADDED
Binary file (2.52 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/meta.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eval/acc_0": 0.8035792308591411,
3
+ "eval/acc_1": 0.6637425632354502,
4
+ "eval/acc_10": 0.18923128577877438,
5
+ "eval/acc_11": 0.17124921611674007,
6
+ "eval/acc_12": 0.15668950770169282,
7
+ "eval/acc_13": 0.1426320067385974,
8
+ "eval/acc_14": 0.13331092176630882,
9
+ "eval/acc_2": 0.5552891023967077,
10
+ "eval/acc_3": 0.46977908434221655,
11
+ "eval/acc_4": 0.4010446327487474,
12
+ "eval/acc_5": 0.3465506263632803,
13
+ "eval/acc_6": 0.30046269263999537,
14
+ "eval/acc_7": 0.26302379392664604,
15
+ "eval/acc_8": 0.23283026018909397,
16
+ "eval/acc_9": 0.20821325150637074,
17
+ "eval/accept_length": 4.327601000587652,
18
+ "eval/accepted_draft_tokens": 3.327601000587652,
19
+ "eval/avg_acc": 0.33626421654404776,
20
+ "eval/avg_loss": 2.461708150162796,
21
+ "eval/block_count": 251730,
22
+ "eval/domain/code/accept_length": 4.6861709411741685,
23
+ "eval/domain/code/accepted_draft_tokens": 3.686170941174169,
24
+ "eval/domain/code/avg_acc": 0.36172285958153183,
25
+ "eval/domain/code/avg_loss": 2.1630422270117546,
26
+ "eval/domain/code/block_count": 51117,
27
+ "eval/domain/code/prompt_count": 100,
28
+ "eval/domain/creative_writing/accept_length": 3.223770442206704,
29
+ "eval/domain/creative_writing/accepted_draft_tokens": 2.223770442206704,
30
+ "eval/domain/creative_writing/avg_acc": 0.2581009412896434,
31
+ "eval/domain/creative_writing/avg_loss": 3.1329529015610498,
32
+ "eval/domain/creative_writing/block_count": 50811,
33
+ "eval/domain/creative_writing/prompt_count": 100,
34
+ "eval/domain/factual_qa/accept_length": 3.470308213651767,
35
+ "eval/domain/factual_qa/accepted_draft_tokens": 2.470308213651767,
36
+ "eval/domain/factual_qa/avg_acc": 0.2759466308226494,
37
+ "eval/domain/factual_qa/avg_loss": 2.8794489069782654,
38
+ "eval/domain/factual_qa/block_count": 49576,
39
+ "eval/domain/factual_qa/prompt_count": 100,
40
+ "eval/domain/general/accept_length": 4.095682850479532,
41
+ "eval/domain/general/accepted_draft_tokens": 3.0956828504795326,
42
+ "eval/domain/general/avg_acc": 0.32111623778680815,
43
+ "eval/domain/general/avg_loss": 2.550857371536131,
44
+ "eval/domain/general/block_count": 49037,
45
+ "eval/domain/general/prompt_count": 100,
46
+ "eval/domain/math/accept_length": 6.117661997694817,
47
+ "eval/domain/math/accepted_draft_tokens": 5.117661997694817,
48
+ "eval/domain/math/avg_acc": 0.4606978048158303,
49
+ "eval/domain/math/avg_loss": 1.6069904789885643,
50
+ "eval/domain/math/block_count": 51189,
51
+ "eval/domain/math/prompt_count": 100,
52
+ "eval/macro5/accept_length": 4.3187188890413974,
53
+ "eval/macro5/accepted_draft_tokens": 3.318718889041398,
54
+ "eval/macro5/avg_acc": 0.3355168948592926,
55
+ "eval/macro5/avg_loss": 2.466658377215153,
56
+ "eval/ploss_0": 0.6291436872532544,
57
+ "eval/ploss_1": 1.2540812805883308,
58
+ "eval/ploss_10": 4.000406688278185,
59
+ "eval/ploss_11": 4.124176352665642,
60
+ "eval/ploss_12": 4.2365122074494455,
61
+ "eval/ploss_13": 4.340265450208499,
62
+ "eval/ploss_14": 4.4291484607792375,
63
+ "eval/ploss_2": 1.8129065321514213,
64
+ "eval/ploss_3": 2.2856744006144503,
65
+ "eval/ploss_4": 2.6804325333015626,
66
+ "eval/ploss_5": 3.0053216023396887,
67
+ "eval/ploss_6": 3.279686727954058,
68
+ "eval/ploss_7": 3.5107537051610183,
69
+ "eval/ploss_8": 3.697533678546502,
70
+ "eval/ploss_9": 3.8625973255129096,
71
+ "eval/prompt_count": 500,
72
+ "eval/wallclock_seconds": 13.282809197437018,
73
+ "global_step": 26650,
74
+ "inference_id": 26650,
75
+ "iteration": 26650,
76
+ "next_inference_id": 26650,
77
+ "next_step": 26650,
78
+ "step": 26650,
79
+ "timestamp": 1792549266.0368798,
80
+ "world_size": 6
81
+ }
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/.metadata ADDED
Binary file (80.4 kB). View file
 
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__0_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2c828d614982ee47533e832f144f751ea168f1a98283adffcac26e4aac6ef56
3
+ size 557189747
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__1_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ed482fafb0f88c08f5f3393bc117e0c64eafa17cfa83199c8ad19df953050ee
3
+ size 557189747
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__2_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3294856bd3d7432df88a962d1421ea8972f2e50651447306434bd0b91bd83aea
3
+ size 557189747
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__3_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c36c5a764acc3b1a916e33888e57062cc7340840376be3609bd4ddd09bd18c3
3
+ size 557189747
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__4_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8546e14a463cf8ab713a18d9c03dcc063cebe46a13602c1a6cd44af93413151
3
+ size 557189747
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/model/__5_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9f8e5dc62b80b1179c6430b45d23d9ed9a413a69cb255e1244d06cf26e2c4e2
3
+ size 556518003
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/.metadata ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c325feb2c142400cb65cfbb8dd9dfa644030de04f7d65b2f7891553fd931c64c
3
+ size 208966
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__0_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f00a68099d2ba1013273241c08f76b831a05b187629229d11e31bf7e158a6fbf
3
+ size 2098182023
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__1_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb0b393356e6b528668e3feca6bff5e267a38f5f7d9fc717169f64613e916747
3
+ size 2098200759
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__2_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae2d971e23a826e6a5f20d9ea166cf407eda4d585fbde8bef824411f84d652a9
3
+ size 2098199370
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__3_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c022eda74080a757bd285b4c03a4360751d42881754840ad5053bef0c2273213
3
+ size 2098202837
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__4_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3ce94bf87259ef9cb747ceddda9431d6d024d0da674effaee7e010eff9485dd
3
+ size 2098202837
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/optimizer/__5_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c61482d531a085064e495c6b0958531c8b20b25a1c675e2d6aa4556239cf0504
3
+ size 2094270677
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbcb2720dee5d972516c5754d5ce56f9478f2052754eb26e04014a37f0afc2bd
3
+ size 14709
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00001.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccd405b0b81e0e45e4855ef9addaff899a58f5e08300b08313a01b2d6cefaef4
3
+ size 14773
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00002.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b251fbc01ef5a6d2a11f3fed60124ee4154da0b530911ff45c633c7b311c5928
3
+ size 14709
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00003.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16ec3b0f56b2f022cea398592faefdacf5fb845c05db320cb044c9da89a42842
3
+ size 14709
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00004.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac8d734471fa1ab45ea0f606725138abf26d41b8d6677d14fae37b2060e31e27
3
+ size 14773
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/rng_rank_00005.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfe987f496df753ca1ec2ba476760eb0b7dcb55d335d8e7259c4c52bb538d75c
3
+ size 14773
e5_8b_800k_20260831/arms/e5-dense-lr2e3/checkpoints/iter_0026650/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": "4f808090f20d17a8ee74e90b25b38e6048f1b4027c12385bee01c9dee9e777e6",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "ff78a938ad3e53f4d6a6666a819648748b2913b52b8c79301cd432750f44da34",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "107dc3e9e15eb49e60fecd2b8f093cf69cc7e2843b69ecbed4ac42039a48e072",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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-lr2e3/checkpoints/iter_0026650/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": "85b74e2149b02bc5734aed768488fb2d83e64bbe9d98bd8f4dfd93db25f3e7a5",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "208cc4a97e2cecaa4b4afd97a6e3ba8ecae4686fa58460cb4c5a58cbfbd3d077",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "ca93c822eadb8eef61e9ecaec9c23a983a781aa414ee490d23b39f6a671cefd6",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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-lr2e3/checkpoints/iter_0026650/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": "4840d92cadec55941bb7b5086b411bffa17216fe5315d0a0ea27226361399410",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "2cf8a640ebdab1b1651eaab8ec91d7462c2634950f44a0fb7e90d8a3c19b8af9",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "3ec6327dc68a82c62d5c878818b7ce8e1aeacd2acd0d110853bd89c139312ed8",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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-lr2e3/checkpoints/iter_0026650/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": "7b5669df7f7fc21c09eefa0a25ef9d3acf5bbde2b90bc4fcf6c41c3abd7459c8",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "9461980ab7327edd01a0add0179a1844b1c34f108360154faa698d068c300c7f",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "66595f46051da981391bf2b3085e45327d17d2babf74ddad2cf713608ec70242",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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-lr2e3/checkpoints/iter_0026650/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": "804f08d025487fdeb836e66b9f75bd8430086ff936dc1cf242af7ce00591436f",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "4773bc4eabcc15c0d1d4c2d6d94214d8899e4dec65702ef20faac4cf57b4d8a8",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "b4af233904cf99dca778c8a08036c9c2b5ad24442072809a7bac0cffed7abd66",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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-lr2e3/checkpoints/iter_0026650/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": "2783acc005b2a299116448ee6919511d68748cc5e94b4e4018ffdc26d05aaf6c",
11
+ "tensor_count": 74
12
+ },
13
+ "optimizer": {
14
+ "sampled_tensor_count": 108,
15
+ "sha256": "5d038b0abc204240433e62dd63b115ac929985027ed4d914ef7b5ed05aaef162",
16
+ "tensor_count": 232
17
+ },
18
+ "rng": {
19
+ "sampled_tensor_count": 0,
20
+ "sha256": "0c3894f2a4872cc0517ca73c8401f7fd198ffcabc9a29a7914811c97f88ce55f",
21
+ "tensor_count": 2
22
+ },
23
+ "scheduler": {
24
+ "sampled_tensor_count": 0,
25
+ "sha256": "3bad53374baea108d39bf8eefddfe2e2d2669d824ee4d12a187333e8ab854dfe",
26
+ "tensor_count": 0
27
+ }
28
+ },
29
+ "global_step": 26650,
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
+ }