jasonfan commited on
Commit
b239e90
·
verified ·
1 Parent(s): 7d7402b

Upload folder using huggingface_hub

Browse files
Files changed (26) hide show
  1. checkpoints/transformer_resume/_checkpoint_log.tsv +223 -222
  2. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000010000_tokens_57.1b_val=0.2374.txt +0 -0
  3. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000011000_tokens_58b_val=0.2374.txt +0 -0
  4. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000012000_tokens_58.8b_val=0.2373.txt +0 -0
  5. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373.txt +0 -0
  6. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank1.txt +0 -0
  7. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank2.txt +0 -0
  8. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank3.txt +0 -0
  9. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank4.txt +0 -0
  10. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank5.txt +0 -0
  11. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank6.txt +0 -0
  12. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank7.txt +0 -0
  13. checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000014000_tokens_60.6b_val=0.2373.txt +0 -0
  14. checkpoints/transformer_resume/metrics_summary_epoch4_top5.csv +1 -1
  15. checkpoints/transformer_resume/verification_epoch_004_step_000010000_tokens_57.1b_val=0.2374.txt +8 -0
  16. checkpoints/transformer_resume/verification_epoch_004_step_000011000_tokens_58b_val=0.2374.txt +5 -0
  17. checkpoints/transformer_resume/verification_epoch_004_step_000012000_tokens_58.8b_val=0.2373.txt +5 -0
  18. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373.txt +5 -0
  19. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank1.txt +5 -0
  20. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank2.txt +5 -0
  21. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank3.txt +5 -0
  22. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank4.txt +5 -0
  23. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank5.txt +5 -0
  24. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank6.txt +5 -0
  25. checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank7.txt +5 -0
  26. checkpoints/transformer_resume/verification_epoch_004_step_000014000_tokens_60.6b_val=0.2373.txt +5 -0
checkpoints/transformer_resume/_checkpoint_log.tsv CHANGED
@@ -1,223 +1,224 @@
1
  Time Checkpoint_Type Epoch Step Train_Loss Val_Loss Tokens_Trained
2
- 2025-12-08 22:04:33 recent 000 000001000 0.6310132890939713 0.5834923386573792 860160000
3
- 2025-12-08 22:04:36 step 000 000001000 0.6310132890939713 0.5834923386573792 860160000
4
- 2025-12-08 22:04:38 best 000 000001000 0.6310132890939713 0.5834923386573792 860160000
5
- 2025-12-08 22:21:23 recent 000 000002000 0.4273352026939392 0.4196219742298126 1720320000
6
- 2025-12-08 22:21:25 step 000 000002000 0.4273352026939392 0.4196219742298126 1720320000
7
- 2025-12-08 22:21:28 best 000 000002000 0.4273352026939392 0.4196219742298126 1720320000
8
- 2025-12-08 22:38:11 recent 000 000003000 0.38565465807914734 0.3750569224357605 2580480000
9
- 2025-12-08 22:38:13 step 000 000003000 0.38565465807914734 0.3750569224357605 2580480000
10
- 2025-12-08 22:38:15 best 000 000003000 0.38565465807914734 0.3750569224357605 2580480000
11
- 2025-12-08 22:54:53 recent 000 000004000 0.37390973418951035 0.3507870137691498 3440640000
12
- 2025-12-08 22:54:55 step 000 000004000 0.37390973418951035 0.3507870137691498 3440640000
13
- 2025-12-08 22:54:57 best 000 000004000 0.37390973418951035 0.3507870137691498 3440640000
14
- 2025-12-08 23:11:41 recent 000 000005000 0.3376680389046669 0.3366142213344574 4300800000
15
- 2025-12-08 23:11:44 step 000 000005000 0.3376680389046669 0.3366142213344574 4300800000
16
- 2025-12-08 23:11:46 best 000 000005000 0.3376680389046669 0.3366142213344574 4300800000
17
- 2025-12-08 23:28:26 recent 000 000006000 0.3262065425515175 0.3260747790336609 5160960000
18
- 2025-12-08 23:28:28 step 000 000006000 0.3262065425515175 0.3260747790336609 5160960000
19
- 2025-12-08 23:28:31 best 000 000006000 0.3262065425515175 0.3260747790336609 5160960000
20
- 2025-12-08 23:45:16 recent 000 000007000 0.30282650142908096 0.31829097867012024 6021120000
21
- 2025-12-08 23:45:18 step 000 000007000 0.30282650142908096 0.31829097867012024 6021120000
22
- 2025-12-08 23:45:21 best 000 000007000 0.30282650142908096 0.31829097867012024 6021120000
23
- 2025-12-09 00:02:05 recent 000 000008000 0.2894744873046875 0.3108223080635071 6881280000
24
- 2025-12-09 00:02:07 step 000 000008000 0.2894744873046875 0.3108223080635071 6881280000
25
- 2025-12-09 00:02:10 best 000 000008000 0.2894744873046875 0.3108223080635071 6881280000
26
- 2025-12-09 00:18:48 recent 000 000009000 0.3270149603486061 0.30675119161605835 7741440000
27
- 2025-12-09 00:18:50 step 000 000009000 0.3270149603486061 0.30675119161605835 7741440000
28
- 2025-12-09 00:18:52 best 000 000009000 0.3270149603486061 0.30675119161605835 7741440000
29
- 2025-12-09 00:35:35 recent 000 000010000 0.3266168162226677 0.30241405963897705 8601600000
30
- 2025-12-09 00:35:38 step 000 000010000 0.3266168162226677 0.30241405963897705 8601600000
31
- 2025-12-09 00:35:40 best 000 000010000 0.3266168162226677 0.30241405963897705 8601600000
32
- 2025-12-09 00:52:23 recent 000 000011000 0.31317316740751266 0.29902809858322144 9461760000
33
- 2025-12-09 00:52:25 step 000 000011000 0.31317316740751266 0.29902809858322144 9461760000
34
- 2025-12-09 00:52:28 best 000 000011000 0.31317316740751266 0.29902809858322144 9461760000
35
- 2025-12-09 01:09:06 recent 000 000012000 0.3402252197265625 0.2953135669231415 10321920000
36
- 2025-12-09 01:09:08 step 000 000012000 0.3402252197265625 0.2953135669231415 10321920000
37
- 2025-12-09 01:09:10 best 000 000012000 0.3402252197265625 0.2953135669231415 10321920000
38
- 2025-12-09 01:25:55 recent 000 000013000 0.29353654384613037 0.29272380471229553 11182080000
39
- 2025-12-09 01:25:57 step 000 000013000 0.29353654384613037 0.29272380471229553 11182080000
40
- 2025-12-09 01:26:00 best 000 000013000 0.29353654384613037 0.29272380471229553 11182080000
41
- 2025-12-09 01:42:39 recent 000 000014000 0.2953661009669304 0.2901206612586975 12042240000
42
- 2025-12-09 01:42:42 step 000 000014000 0.2953661009669304 0.2901206612586975 12042240000
43
- 2025-12-09 01:42:45 best 000 000014000 0.2953661009669304 0.2901206612586975 12042240000
44
- 2025-12-09 01:45:35 recent 001 000000000 0.31282012164592743 0.2897591292858124 12129976320
45
- 2025-12-09 01:45:37 epoch 001 000000000 0.31282012164592743 0.2897591292858124 12129976320
46
- 2025-12-09 01:45:40 best 001 000000000 0.31282012164592743 0.2897591292858124 12129976320
47
- 2025-12-09 02:02:21 recent 001 000001000 0.29584021121263504 0.2877645790576935 12990136320
48
- 2025-12-09 02:02:23 step 001 000001000 0.29584021121263504 0.2877645790576935 12990136320
49
- 2025-12-09 02:02:26 best 001 000001000 0.29584021121263504 0.2877645790576935 12990136320
50
- 2025-12-09 02:19:10 recent 001 000002000 0.2957344278693199 0.28560930490493774 13850296320
51
- 2025-12-09 02:19:13 step 001 000002000 0.2957344278693199 0.28560930490493774 13850296320
52
- 2025-12-09 02:19:16 best 001 000002000 0.2957344278693199 0.28560930490493774 13850296320
53
- 2025-12-09 02:35:54 recent 001 000003000 0.2833305895328522 0.2839827239513397 14710456320
54
- 2025-12-09 02:35:56 step 001 000003000 0.2833305895328522 0.2839827239513397 14710456320
55
- 2025-12-09 02:35:59 best 001 000003000 0.2833305895328522 0.2839827239513397 14710456320
56
- 2025-12-09 02:52:43 recent 001 000004000 0.30934131145477295 0.28247708082199097 15570616320
57
- 2025-12-09 02:52:46 step 001 000004000 0.30934131145477295 0.28247708082199097 15570616320
58
- 2025-12-09 02:52:49 best 001 000004000 0.30934131145477295 0.28247708082199097 15570616320
59
- 2025-12-09 03:09:33 recent 001 000005000 0.2961928993463516 0.28070640563964844 16430776320
60
- 2025-12-09 03:09:35 step 001 000005000 0.2961928993463516 0.28070640563964844 16430776320
61
- 2025-12-09 03:09:38 best 001 000005000 0.2961928993463516 0.28070640563964844 16430776320
62
- 2025-12-09 03:26:17 recent 001 000006000 0.274027444422245 0.279300332069397 17290936320
63
- 2025-12-09 03:26:19 step 001 000006000 0.274027444422245 0.279300332069397 17290936320
64
- 2025-12-09 03:26:22 best 001 000006000 0.274027444422245 0.279300332069397 17290936320
65
- 2025-12-09 03:43:04 recent 001 000007000 0.29656217247247696 0.27778464555740356 18151096320
66
- 2025-12-09 03:43:06 step 001 000007000 0.29656217247247696 0.27778464555740356 18151096320
67
- 2025-12-09 03:43:09 best 001 000007000 0.29656217247247696 0.27778464555740356 18151096320
68
- 2025-12-09 03:59:54 recent 001 000008000 0.3060613349080086 0.27626287937164307 19011256320
69
- 2025-12-09 03:59:56 step 001 000008000 0.3060613349080086 0.27626287937164307 19011256320
70
- 2025-12-09 03:59:59 best 001 000008000 0.3060613349080086 0.27626287937164307 19011256320
71
- 2025-12-09 04:16:37 recent 001 000009000 0.26544981449842453 0.2751830816268921 19871416320
72
- 2025-12-09 04:16:39 step 001 000009000 0.26544981449842453 0.2751830816268921 19871416320
73
- 2025-12-09 04:16:41 best 001 000009000 0.26544981449842453 0.2751830816268921 19871416320
74
- 2025-12-09 04:33:26 recent 001 000010000 0.30200086534023285 0.27379918098449707 20731576320
75
- 2025-12-09 04:33:29 step 001 000010000 0.30200086534023285 0.27379918098449707 20731576320
76
- 2025-12-09 04:33:31 best 001 000010000 0.30200086534023285 0.27379918098449707 20731576320
77
- 2025-12-09 04:50:11 recent 001 000011000 0.2979862913489342 0.27259981632232666 21591736320
78
- 2025-12-09 04:50:14 step 001 000011000 0.2979862913489342 0.27259981632232666 21591736320
79
- 2025-12-09 04:50:16 best 001 000011000 0.2979862913489342 0.27259981632232666 21591736320
80
- 2025-12-09 05:07:02 recent 001 000012000 0.2857537269592285 0.27134189009666443 22451896320
81
- 2025-12-09 05:07:04 step 001 000012000 0.2857537269592285 0.27134189009666443 22451896320
82
- 2025-12-09 05:07:07 best 001 000012000 0.2857537269592285 0.27134189009666443 22451896320
83
- 2025-12-09 05:23:52 recent 001 000013000 0.2915356382727623 0.2702634632587433 23312056320
84
- 2025-12-09 05:23:54 step 001 000013000 0.2915356382727623 0.2702634632587433 23312056320
85
- 2025-12-09 05:23:57 best 001 000013000 0.2915356382727623 0.2702634632587433 23312056320
86
- 2025-12-09 05:40:36 recent 001 000014000 0.2914014160633087 0.26911184191703796 24172216320
87
- 2025-12-09 05:40:38 step 001 000014000 0.2914014160633087 0.26911184191703796 24172216320
88
- 2025-12-09 05:40:41 best 001 000014000 0.2914014160633087 0.26911184191703796 24172216320
89
- 2025-12-09 05:43:19 recent 002 000000000 0.29752761870622635 0.2691074013710022 24259952640
90
- 2025-12-09 05:43:21 epoch 002 000000000 0.29752761870622635 0.2691074013710022 24259952640
91
- 2025-12-09 05:43:24 best 002 000000000 0.29752761870622635 0.2691074013710022 24259952640
92
- 2025-12-09 06:00:11 recent 002 000001000 0.2540785148739815 0.2674492299556732 25120112640
93
- 2025-12-09 06:00:13 step 002 000001000 0.2540785148739815 0.2674492299556732 25120112640
94
- 2025-12-09 06:00:15 best 002 000001000 0.2540785148739815 0.2674492299556732 25120112640
95
- 2025-12-09 06:17:00 recent 002 000002000 0.2908788323402405 0.26671120524406433 25980272640
96
- 2025-12-09 06:17:03 step 002 000002000 0.2908788323402405 0.26671120524406433 25980272640
97
- 2025-12-09 06:17:05 best 002 000002000 0.2908788323402405 0.26671120524406433 25980272640
98
- 2025-12-09 06:33:46 recent 002 000003000 0.26616644859313965 0.26544100046157837 26840432640
99
- 2025-12-09 06:33:48 step 002 000003000 0.26616644859313965 0.26544100046157837 26840432640
100
- 2025-12-09 06:33:51 best 002 000003000 0.26616644859313965 0.26544100046157837 26840432640
101
- 2025-12-09 06:50:35 recent 002 000004000 0.26277821511030197 0.2647247016429901 27700592640
102
- 2025-12-09 06:50:37 step 002 000004000 0.26277821511030197 0.2647247016429901 27700592640
103
- 2025-12-09 06:50:40 best 002 000004000 0.26277821511030197 0.2647247016429901 27700592640
104
- 2025-12-09 07:07:21 recent 002 000005000 0.2830668017268181 0.26385506987571716 28560752640
105
- 2025-12-09 07:07:23 step 002 000005000 0.2830668017268181 0.26385506987571716 28560752640
106
- 2025-12-09 07:07:25 best 002 000005000 0.2830668017268181 0.26385506987571716 28560752640
107
- 2025-12-09 07:24:06 recent 002 000006000 0.2711658403277397 0.26265472173690796 29420912640
108
- 2025-12-09 07:24:08 step 002 000006000 0.2711658403277397 0.26265472173690796 29420912640
109
- 2025-12-09 07:24:10 best 002 000006000 0.2711658403277397 0.26265472173690796 29420912640
110
- 2025-12-09 07:40:55 recent 002 000007000 0.27206581085920334 0.26172950863838196 30281072640
111
- 2025-12-09 07:40:58 step 002 000007000 0.27206581085920334 0.26172950863838196 30281072640
112
- 2025-12-09 07:41:01 best 002 000007000 0.27206581085920334 0.26172950863838196 30281072640
113
- 2025-12-09 07:57:41 recent 002 000008000 0.26128602772951126 0.26091468334198 31141232640
114
- 2025-12-09 07:57:43 step 002 000008000 0.26128602772951126 0.26091468334198 31141232640
115
- 2025-12-09 07:57:46 best 002 000008000 0.26128602772951126 0.26091468334198 31141232640
116
- 2025-12-09 08:14:30 recent 002 000009000 0.2547329440712929 0.25993025302886963 32001392640
117
- 2025-12-09 08:14:32 step 002 000009000 0.2547329440712929 0.25993025302886963 32001392640
118
- 2025-12-09 08:14:35 best 002 000009000 0.2547329440712929 0.25993025302886963 32001392640
119
- 2025-12-09 08:31:20 recent 002 000010000 0.26767709851264954 0.2591801583766937 32861552640
120
- 2025-12-09 08:31:22 step 002 000010000 0.26767709851264954 0.2591801583766937 32861552640
121
- 2025-12-09 08:31:25 best 002 000010000 0.26767709851264954 0.2591801583766937 32861552640
122
- 2025-12-09 08:48:03 recent 002 000011000 0.2763865441083908 0.25831955671310425 33721712640
123
- 2025-12-09 08:48:06 step 002 000011000 0.2763865441083908 0.25831955671310425 33721712640
124
- 2025-12-09 08:48:08 best 002 000011000 0.2763865441083908 0.25831955671310425 33721712640
125
- 2025-12-09 09:04:52 recent 002 000012000 0.26893263310194016 0.25764235854148865 34581872640
126
- 2025-12-09 09:04:54 step 002 000012000 0.26893263310194016 0.25764235854148865 34581872640
127
- 2025-12-09 09:04:56 best 002 000012000 0.26893263310194016 0.25764235854148865 34581872640
128
- 2025-12-09 09:21:38 recent 002 000013000 0.2613510861992836 0.2566717565059662 35442032640
129
- 2025-12-09 09:21:41 step 002 000013000 0.2613510861992836 0.2566717565059662 35442032640
130
- 2025-12-09 09:21:43 best 002 000013000 0.2613510861992836 0.2566717565059662 35442032640
131
- 2025-12-09 09:38:24 recent 002 000014000 0.2777568846940994 0.25575369596481323 36302192640
132
- 2025-12-09 09:38:26 step 002 000014000 0.2777568846940994 0.25575369596481323 36302192640
133
- 2025-12-09 09:38:28 best 002 000014000 0.2777568846940994 0.25575369596481323 36302192640
134
- 2025-12-09 09:41:05 recent 003 000000000 0.25520671159029007 0.2558022141456604 36389928960
135
- 2025-12-09 09:41:07 epoch 003 000000000 0.25520671159029007 0.2558022141456604 36389928960
136
- 2025-12-09 09:57:55 recent 003 000001000 0.26569607853889465 0.25445839762687683 37250088960
137
- 2025-12-09 09:57:58 step 003 000001000 0.26569607853889465 0.25445839762687683 37250088960
138
- 2025-12-09 09:58:00 best 003 000001000 0.26569607853889465 0.25445839762687683 37250088960
139
- 2025-12-09 10:14:41 recent 003 000002000 0.2589205652475357 0.2537902593612671 38110248960
140
- 2025-12-09 10:14:43 step 003 000002000 0.2589205652475357 0.2537902593612671 38110248960
141
- 2025-12-09 10:14:46 best 003 000002000 0.2589205652475357 0.2537902593612671 38110248960
142
- 2025-12-09 10:31:29 recent 003 000003000 0.2700107917189598 0.2531486451625824 38970408960
143
- 2025-12-09 10:31:32 step 003 000003000 0.2700107917189598 0.2531486451625824 38970408960
144
- 2025-12-09 10:31:34 best 003 000003000 0.2700107917189598 0.2531486451625824 38970408960
145
- 2025-12-09 10:48:19 recent 003 000004000 0.2698415145277977 0.2523936629295349 39830568960
146
- 2025-12-09 10:48:21 step 003 000004000 0.2698415145277977 0.2523936629295349 39830568960
147
- 2025-12-09 10:48:24 best 003 000004000 0.2698415145277977 0.2523936629295349 39830568960
148
- 2025-12-09 11:05:04 recent 003 000005000 0.2487308233976364 0.25164473056793213 40690728960
149
- 2025-12-09 11:05:07 step 003 000005000 0.2487308233976364 0.25164473056793213 40690728960
150
- 2025-12-09 11:05:10 best 003 000005000 0.2487308233976364 0.25164473056793213 40690728960
151
- 2025-12-09 11:21:52 recent 003 000006000 0.25990080088377 0.2509293556213379 41550888960
152
- 2025-12-09 11:21:54 step 003 000006000 0.25990080088377 0.2509293556213379 41550888960
153
- 2025-12-09 11:21:56 best 003 000006000 0.25990080088377 0.2509293556213379 41550888960
154
- 2025-12-09 11:38:42 recent 003 000007000 0.26555294543504715 0.2501857578754425 42411048960
155
- 2025-12-09 11:38:44 step 003 000007000 0.26555294543504715 0.2501857578754425 42411048960
156
- 2025-12-09 11:38:47 best 003 000007000 0.26555294543504715 0.2501857578754425 42411048960
157
- 2025-12-09 11:55:27 recent 003 000008000 0.27267616987228394 0.24950169026851654 43271208960
158
- 2025-12-09 11:55:30 step 003 000008000 0.27267616987228394 0.24950169026851654 43271208960
159
- 2025-12-09 11:55:32 best 003 000008000 0.27267616987228394 0.24950169026851654 43271208960
160
- 2025-12-09 12:12:16 recent 003 000009000 0.2616996392607689 0.24902282655239105 44131368960
161
- 2025-12-09 12:12:18 step 003 000009000 0.2616996392607689 0.24902282655239105 44131368960
162
- 2025-12-09 12:12:21 best 003 000009000 0.2616996392607689 0.24902282655239105 44131368960
163
- 2025-12-09 12:29:00 recent 003 000010000 0.2559799402952194 0.24838656187057495 44991528960
164
- 2025-12-09 12:29:02 step 003 000010000 0.2559799402952194 0.24838656187057495 44991528960
165
- 2025-12-09 12:29:04 best 003 000010000 0.2559799402952194 0.24838656187057495 44991528960
166
- 2025-12-09 12:45:48 recent 003 000011000 0.2760173827409744 0.24766436219215393 45851688960
167
- 2025-12-09 12:45:51 step 003 000011000 0.2760173827409744 0.24766436219215393 45851688960
168
- 2025-12-09 12:45:53 best 003 000011000 0.2760173827409744 0.24766436219215393 45851688960
169
- 2025-12-09 13:02:38 recent 003 000012000 0.2683108448982239 0.24721525609493256 46711848960
170
- 2025-12-09 13:02:40 step 003 000012000 0.2683108448982239 0.24721525609493256 46711848960
171
- 2025-12-09 13:02:43 best 003 000012000 0.2683108448982239 0.24721525609493256 46711848960
172
- 2025-12-09 13:19:22 recent 003 000013000 0.25255977362394333 0.24673430621623993 47572008960
173
- 2025-12-09 13:19:24 step 003 000013000 0.25255977362394333 0.24673430621623993 47572008960
174
- 2025-12-09 13:19:27 best 003 000013000 0.25255977362394333 0.24673430621623993 47572008960
175
- 2025-12-09 13:36:11 recent 003 000014000 0.2679848149418831 0.24626043438911438 48432168960
176
- 2025-12-09 13:36:13 step 003 000014000 0.2679848149418831 0.24626043438911438 48432168960
177
- 2025-12-09 13:36:15 best 003 000014000 0.2679848149418831 0.24626043438911438 48432168960
178
- 2025-12-09 13:38:55 recent 004 000000000 0.25628816336393356 0.24614378809928894 48519905280
179
- 2025-12-09 13:38:57 epoch 004 000000000 0.25628816336393356 0.24614378809928894 48519905280
180
- 2025-12-09 13:39:00 best 004 000000000 0.25628816336393356 0.24614378809928894 48519905280
181
- 2025-12-09 13:55:47 recent 004 000001000 0.24197009950876236 0.24566997587680817 49380065280
182
- 2025-12-09 13:55:50 step 004 000001000 0.24197009950876236 0.24566997587680817 49380065280
183
- 2025-12-09 13:55:52 best 004 000001000 0.24197009950876236 0.24566997587680817 49380065280
184
- 2025-12-09 14:12:33 recent 004 000002000 0.2547924369573593 0.24531063437461853 50240225280
185
- 2025-12-09 14:12:35 step 004 000002000 0.2547924369573593 0.24531063437461853 50240225280
186
- 2025-12-09 14:12:38 best 004 000002000 0.2547924369573593 0.24531063437461853 50240225280
187
- 2025-12-09 14:29:22 recent 004 000003000 0.2657608091831207 0.2449333518743515 51100385280
188
- 2025-12-09 14:29:24 step 004 000003000 0.2657608091831207 0.2449333518743515 51100385280
189
- 2025-12-09 14:29:26 best 004 000003000 0.2657608091831207 0.2449333518743515 51100385280
190
- 2025-12-09 14:46:11 recent 004 000004000 0.2466345801949501 0.24462857842445374 51960545280
191
- 2025-12-09 14:46:13 step 004 000004000 0.2466345801949501 0.24462857842445374 51960545280
192
- 2025-12-09 14:46:16 best 004 000004000 0.2466345801949501 0.24462857842445374 51960545280
193
- 2025-12-09 15:02:57 recent 004 000005000 0.26587440073490143 0.24436171352863312 52820705280
194
- 2025-12-09 15:02:59 step 004 000005000 0.26587440073490143 0.24436171352863312 52820705280
195
- 2025-12-09 15:03:02 best 004 000005000 0.26587440073490143 0.24436171352863312 52820705280
196
- 2025-12-09 15:19:46 recent 004 000006000 0.24569455534219742 0.2440834641456604 53680865280
197
- 2025-12-09 15:19:49 step 004 000006000 0.24569455534219742 0.2440834641456604 53680865280
198
- 2025-12-09 15:19:51 best 004 000006000 0.24569455534219742 0.2440834641456604 53680865280
199
- 2025-12-09 15:36:32 recent 004 000007000 0.24603911489248276 0.24390318989753723 54541025280
200
- 2025-12-09 15:36:34 step 004 000007000 0.24603911489248276 0.24390318989753723 54541025280
201
- 2025-12-09 15:36:37 best 004 000007000 0.24603911489248276 0.24390318989753723 54541025280
202
- 2025-12-09 15:53:21 recent 004 000008000 0.2595767304301262 0.24374566972255707 55401185280
203
- 2025-12-09 15:53:23 step 004 000008000 0.2595767304301262 0.24374566972255707 55401185280
204
- 2025-12-09 15:53:25 best 004 000008000 0.2595767304301262 0.24374566972255707 55401185280
205
- 2025-12-09 16:10:10 recent 004 000009000 0.2663945034146309 0.24360939860343933 56261345280
206
- 2025-12-09 16:10:12 step 004 000009000 0.2663945034146309 0.24360939860343933 56261345280
207
- 2025-12-09 16:10:14 best 004 000009000 0.2663945034146309 0.24360939860343933 56261345280
208
- 2025-12-09 16:26:53 recent 004 000010000 0.24528992921113968 0.24356481432914734 57121505280
209
- 2025-12-09 16:26:56 step 004 000010000 0.24528992921113968 0.24356481432914734 57121505280
210
- 2025-12-09 16:26:58 best 004 000010000 0.24528992921113968 0.24356481432914734 57121505280
211
- 2025-12-09 16:43:42 recent 004 000011000 0.2570103481411934 0.24347716569900513 57981665280
212
- 2025-12-09 16:43:44 step 004 000011000 0.2570103481411934 0.24347716569900513 57981665280
213
- 2025-12-09 16:43:47 best 004 000011000 0.2570103481411934 0.24347716569900513 57981665280
214
- 2025-12-09 17:00:31 recent 004 000012000 0.25251463800668716 0.24342156946659088 58841825280
215
- 2025-12-09 17:00:33 step 004 000012000 0.25251463800668716 0.24342156946659088 58841825280
216
- 2025-12-09 17:00:36 best 004 000012000 0.25251463800668716 0.24342156946659088 58841825280
217
- 2025-12-09 17:17:15 recent 004 000013000 0.2467445805668831 0.24339288473129272 59701985280
218
- 2025-12-09 17:17:17 step 004 000013000 0.2467445805668831 0.24339288473129272 59701985280
219
- 2025-12-09 17:17:20 best 004 000013000 0.2467445805668831 0.24339288473129272 59701985280
220
- 2025-12-09 17:34:06 recent 004 000014000 0.24387464672327042 0.24340687692165375 60562145280
221
- 2025-12-09 17:34:08 step 004 000014000 0.24387464672327042 0.24340687692165375 60562145280
222
- 2025-12-09 17:36:48 recent 005 000000000 0.2529599890112877 0.2434108555316925 60649881600
223
- 2025-12-09 17:36:50 epoch 005 000000000 0.2529599890112877 0.2434108555316925 60649881600
 
 
1
  Time Checkpoint_Type Epoch Step Train_Loss Val_Loss Tokens_Trained
2
+ 2025-12-11 12:04:15 recent 000 000001000 0.3439227119088173 0.3244720697402954 860160000
3
+ 2025-12-11 12:04:16 step 000 000001000 0.3439227119088173 0.3244720697402954 860160000
4
+ 2025-12-11 12:04:18 best 000 000001000 0.3439227119088173 0.3244720697402954 860160000
5
+ 2025-12-11 12:18:46 recent 000 000002000 0.30961067974567413 0.3051096796989441 1720320000
6
+ 2025-12-11 12:18:48 step 000 000002000 0.30961067974567413 0.3051096796989441 1720320000
7
+ 2025-12-11 12:18:49 best 000 000002000 0.30961067974567413 0.3051096796989441 1720320000
8
+ 2025-12-11 12:33:26 recent 000 000003000 0.3056980222463608 0.29670092463493347 2580480000
9
+ 2025-12-11 12:33:28 step 000 000003000 0.3056980222463608 0.29670092463493347 2580480000
10
+ 2025-12-11 12:33:30 best 000 000003000 0.3056980222463608 0.29670092463493347 2580480000
11
+ 2025-12-11 12:48:06 recent 000 000004000 0.3108784258365631 0.29011571407318115 3440640000
12
+ 2025-12-11 12:48:07 step 000 000004000 0.3108784258365631 0.29011571407318115 3440640000
13
+ 2025-12-11 12:48:09 best 000 000004000 0.3108784258365631 0.29011571407318115 3440640000
14
+ 2025-12-11 13:02:43 recent 000 000005000 0.2862788438796997 0.2865026295185089 4300800000
15
+ 2025-12-11 13:02:45 step 000 000005000 0.2862788438796997 0.2865026295185089 4300800000
16
+ 2025-12-11 13:02:47 best 000 000005000 0.2862788438796997 0.2865026295185089 4300800000
17
+ 2025-12-11 13:17:18 recent 000 000006000 0.2902119755744934 0.2830173671245575 5160960000
18
+ 2025-12-11 13:17:19 step 000 000006000 0.2902119755744934 0.2830173671245575 5160960000
19
+ 2025-12-11 13:17:21 best 000 000006000 0.2902119755744934 0.2830173671245575 5160960000
20
+ 2025-12-11 13:31:52 recent 000 000007000 0.2733738273382187 0.2810177505016327 6021120000
21
+ 2025-12-11 13:31:54 step 000 000007000 0.2733738273382187 0.2810177505016327 6021120000
22
+ 2025-12-11 13:31:55 best 000 000007000 0.2733738273382187 0.2810177505016327 6021120000
23
+ 2025-12-11 13:46:31 recent 000 000008000 0.2640279680490494 0.2785384953022003 6881280000
24
+ 2025-12-11 13:46:33 step 000 000008000 0.2640279680490494 0.2785384953022003 6881280000
25
+ 2025-12-11 13:46:35 best 000 000008000 0.2640279680490494 0.2785384953022003 6881280000
26
+ 2025-12-11 14:01:10 recent 000 000009000 0.29455964267253876 0.27658748626708984 7741440000
27
+ 2025-12-11 14:01:12 step 000 000009000 0.29455964267253876 0.27658748626708984 7741440000
28
+ 2025-12-11 14:01:13 best 000 000009000 0.29455964267253876 0.27658748626708984 7741440000
29
+ 2025-12-11 14:15:49 recent 000 000010000 0.2957393601536751 0.2751111090183258 8601600000
30
+ 2025-12-11 14:15:51 step 000 000010000 0.2957393601536751 0.2751111090183258 8601600000
31
+ 2025-12-11 14:15:52 best 000 000010000 0.2957393601536751 0.2751111090183258 8601600000
32
+ 2025-12-11 14:30:22 recent 000 000011000 0.2893889471888542 0.27389711141586304 9461760000
33
+ 2025-12-11 14:30:23 step 000 000011000 0.2893889471888542 0.27389711141586304 9461760000
34
+ 2025-12-11 14:30:25 best 000 000011000 0.2893889471888542 0.27389711141586304 9461760000
35
+ 2025-12-11 14:45:00 recent 000 000012000 0.31310197710990906 0.2722736597061157 10321920000
36
+ 2025-12-11 14:45:02 step 000 000012000 0.31310197710990906 0.2722736597061157 10321920000
37
+ 2025-12-11 14:45:03 best 000 000012000 0.31310197710990906 0.2722736597061157 10321920000
38
+ 2025-12-11 14:59:39 recent 000 000013000 0.27358604967594147 0.2708395719528198 11182080000
39
+ 2025-12-11 14:59:41 step 000 000013000 0.27358604967594147 0.2708395719528198 11182080000
40
+ 2025-12-11 14:59:42 best 000 000013000 0.27358604967594147 0.2708395719528198 11182080000
41
+ 2025-12-11 15:14:17 recent 000 000014000 0.2762528583407402 0.2696157395839691 12042240000
42
+ 2025-12-11 15:14:19 step 000 000014000 0.2762528583407402 0.2696157395839691 12042240000
43
+ 2025-12-11 15:14:21 best 000 000014000 0.2762528583407402 0.2696157395839691 12042240000
44
+ 2025-12-11 15:16:39 recent 001 000000000 0.2907368093729019 0.2694391906261444 12129976320
45
+ 2025-12-11 15:16:41 epoch 001 000000000 0.2907368093729019 0.2694391906261444 12129976320
46
+ 2025-12-11 15:16:42 best 001 000000000 0.2907368093729019 0.2694391906261444 12129976320
47
+ 2025-12-11 15:31:19 recent 001 000001000 0.276254840195179 0.2685832977294922 12990136320
48
+ 2025-12-11 15:31:20 step 001 000001000 0.276254840195179 0.2685832977294922 12990136320
49
+ 2025-12-11 15:31:22 best 001 000001000 0.276254840195179 0.2685832977294922 12990136320
50
+ 2025-12-11 15:45:51 recent 001 000002000 0.27662599086761475 0.2673291862010956 13850296320
51
+ 2025-12-11 15:45:52 step 001 000002000 0.27662599086761475 0.2673291862010956 13850296320
52
+ 2025-12-11 15:45:54 best 001 000002000 0.27662599086761475 0.2673291862010956 13850296320
53
+ 2025-12-11 16:00:28 recent 001 000003000 0.26816731691360474 0.2670857310295105 14710456320
54
+ 2025-12-11 16:00:29 step 001 000003000 0.26816731691360474 0.2670857310295105 14710456320
55
+ 2025-12-11 16:00:31 best 001 000003000 0.26816731691360474 0.2670857310295105 14710456320
56
+ 2025-12-11 16:15:06 recent 001 000004000 0.2897113338112831 0.2655026614665985 15570616320
57
+ 2025-12-11 16:15:08 step 001 000004000 0.2897113338112831 0.2655026614665985 15570616320
58
+ 2025-12-11 16:15:09 best 001 000004000 0.2897113338112831 0.2655026614665985 15570616320
59
+ 2025-12-11 16:29:45 recent 001 000005000 0.27645236253738403 0.2644156217575073 16430776320
60
+ 2025-12-11 16:29:47 step 001 000005000 0.27645236253738403 0.2644156217575073 16430776320
61
+ 2025-12-11 16:29:48 best 001 000005000 0.27645236253738403 0.2644156217575073 16430776320
62
+ 2025-12-11 16:44:19 recent 001 000006000 0.25979746878147125 0.26352187991142273 17290936320
63
+ 2025-12-11 16:44:21 step 001 000006000 0.25979746878147125 0.26352187991142273 17290936320
64
+ 2025-12-11 16:44:22 best 001 000006000 0.25979746878147125 0.26352187991142273 17290936320
65
+ 2025-12-11 16:58:55 recent 001 000007000 0.2774749845266342 0.2626994550228119 18151096320
66
+ 2025-12-11 16:58:57 step 001 000007000 0.2774749845266342 0.2626994550228119 18151096320
67
+ 2025-12-11 16:58:59 best 001 000007000 0.2774749845266342 0.2626994550228119 18151096320
68
+ 2025-12-11 17:13:33 recent 001 000008000 0.2860962301492691 0.2617444396018982 19011256320
69
+ 2025-12-11 17:13:35 step 001 000008000 0.2860962301492691 0.2617444396018982 19011256320
70
+ 2025-12-11 17:13:36 best 001 000008000 0.2860962301492691 0.2617444396018982 19011256320
71
+ 2025-12-11 17:28:13 recent 001 000009000 0.25598638504743576 0.26100030541419983 19871416320
72
+ 2025-12-11 17:28:14 step 001 000009000 0.25598638504743576 0.26100030541419983 19871416320
73
+ 2025-12-11 17:28:16 best 001 000009000 0.25598638504743576 0.26100030541419983 19871416320
74
+ 2025-12-11 17:42:52 recent 001 000010000 0.28342578560113907 0.26024317741394043 20731576320
75
+ 2025-12-11 17:42:53 step 001 000010000 0.28342578560113907 0.26024317741394043 20731576320
76
+ 2025-12-11 17:42:55 best 001 000010000 0.28342578560113907 0.26024317741394043 20731576320
77
+ 2025-12-11 17:57:24 recent 001 000011000 0.2861173525452614 0.259305864572525 21591736320
78
+ 2025-12-11 17:57:26 step 001 000011000 0.2861173525452614 0.259305864572525 21591736320
79
+ 2025-12-11 17:57:28 best 001 000011000 0.2861173525452614 0.259305864572525 21591736320
80
+ 2025-12-11 18:12:03 recent 001 000012000 0.2670879513025284 0.25852206349372864 22451896320
81
+ 2025-12-11 18:12:05 step 001 000012000 0.2670879513025284 0.25852206349372864 22451896320
82
+ 2025-12-11 18:12:06 best 001 000012000 0.2670879513025284 0.25852206349372864 22451896320
83
+ 2025-12-11 18:26:42 recent 001 000013000 0.2744786664843559 0.2578280568122864 23312056320
84
+ 2025-12-11 18:26:43 step 001 000013000 0.2744786664843559 0.2578280568122864 23312056320
85
+ 2025-12-11 18:26:45 best 001 000013000 0.2744786664843559 0.2578280568122864 23312056320
86
+ 2025-12-11 18:41:21 recent 001 000014000 0.2746414467692375 0.2572340965270996 24172216320
87
+ 2025-12-11 18:41:23 step 001 000014000 0.2746414467692375 0.2572340965270996 24172216320
88
+ 2025-12-11 18:41:24 best 001 000014000 0.2746414467692375 0.2572340965270996 24172216320
89
+ 2025-12-11 18:43:34 recent 002 000000000 0.27967356890439987 0.25697261095046997 24259952640
90
+ 2025-12-11 18:43:35 epoch 002 000000000 0.27967356890439987 0.25697261095046997 24259952640
91
+ 2025-12-11 18:43:37 best 002 000000000 0.27967356890439987 0.25697261095046997 24259952640
92
+ 2025-12-11 18:58:13 recent 002 000001000 0.24822090566158295 0.25578463077545166 25120112640
93
+ 2025-12-11 18:58:15 step 002 000001000 0.24822090566158295 0.25578463077545166 25120112640
94
+ 2025-12-11 18:58:16 best 002 000001000 0.24822090566158295 0.25578463077545166 25120112640
95
+ 2025-12-11 19:12:46 recent 002 000002000 0.2794698253273964 0.25524935126304626 25980272640
96
+ 2025-12-11 19:12:47 step 002 000002000 0.2794698253273964 0.25524935126304626 25980272640
97
+ 2025-12-11 19:12:49 best 002 000002000 0.2794698253273964 0.25524935126304626 25980272640
98
+ 2025-12-11 19:27:24 recent 002 000003000 0.25624027103185654 0.2544611990451813 26840432640
99
+ 2025-12-11 19:27:25 step 002 000003000 0.25624027103185654 0.2544611990451813 26840432640
100
+ 2025-12-11 19:27:27 best 002 000003000 0.25624027103185654 0.2544611990451813 26840432640
101
+ 2025-12-11 19:42:03 recent 002 000004000 0.25411998480558395 0.25379055738449097 27700592640
102
+ 2025-12-11 19:42:04 step 002 000004000 0.25411998480558395 0.25379055738449097 27700592640
103
+ 2025-12-11 19:42:06 best 002 000004000 0.25411998480558395 0.25379055738449097 27700592640
104
+ 2025-12-11 19:56:41 recent 002 000005000 0.27206335216760635 0.25304773449897766 28560752640
105
+ 2025-12-11 19:56:43 step 002 000005000 0.27206335216760635 0.25304773449897766 28560752640
106
+ 2025-12-11 19:56:45 best 002 000005000 0.27206335216760635 0.25304773449897766 28560752640
107
+ 2025-12-11 20:11:14 recent 002 000006000 0.259377621114254 0.25215402245521545 29420912640
108
+ 2025-12-11 20:11:16 step 002 000006000 0.259377621114254 0.25215402245521545 29420912640
109
+ 2025-12-11 20:11:18 best 002 000006000 0.259377621114254 0.25215402245521545 29420912640
110
+ 2025-12-11 20:25:54 recent 002 000007000 0.259706512093544 0.251708060503006 30281072640
111
+ 2025-12-11 20:25:56 step 002 000007000 0.259706512093544 0.251708060503006 30281072640
112
+ 2025-12-11 20:25:57 best 002 000007000 0.259706512093544 0.251708060503006 30281072640
113
+ 2025-12-11 20:40:32 recent 002 000008000 0.2519996836781502 0.251171350479126 31141232640
114
+ 2025-12-11 20:40:34 step 002 000008000 0.2519996836781502 0.251171350479126 31141232640
115
+ 2025-12-11 20:40:36 best 002 000008000 0.2519996836781502 0.251171350479126 31141232640
116
+ 2025-12-11 20:55:11 recent 002 000009000 0.2465052455663681 0.2502414286136627 32001392640
117
+ 2025-12-11 20:55:12 step 002 000009000 0.2465052455663681 0.2502414286136627 32001392640
118
+ 2025-12-11 20:55:14 best 002 000009000 0.2465052455663681 0.2502414286136627 32001392640
119
+ 2025-12-11 21:09:51 recent 002 000010000 0.2524038702249527 0.24950461089611053 32861552640
120
+ 2025-12-11 21:09:52 step 002 000010000 0.2524038702249527 0.24950461089611053 32861552640
121
+ 2025-12-11 21:09:54 best 002 000010000 0.2524038702249527 0.24950461089611053 32861552640
122
+ 2025-12-11 21:24:23 recent 002 000011000 0.2638489082455635 0.2491004317998886 33721712640
123
+ 2025-12-11 21:24:24 step 002 000011000 0.2638489082455635 0.2491004317998886 33721712640
124
+ 2025-12-11 21:24:26 best 002 000011000 0.2638489082455635 0.2491004317998886 33721712640
125
+ 2025-12-11 21:39:01 recent 002 000012000 0.2622009664773941 0.24844156205654144 34581872640
126
+ 2025-12-11 21:39:03 step 002 000012000 0.2622009664773941 0.24844156205654144 34581872640
127
+ 2025-12-11 21:39:05 best 002 000012000 0.2622009664773941 0.24844156205654144 34581872640
128
+ 2025-12-11 21:53:40 recent 002 000013000 0.25010908395051956 0.2477104663848877 35442032640
129
+ 2025-12-11 21:53:41 step 002 000013000 0.25010908395051956 0.2477104663848877 35442032640
130
+ 2025-12-11 21:53:43 best 002 000013000 0.25010908395051956 0.2477104663848877 35442032640
131
+ 2025-12-11 22:08:17 recent 002 000014000 0.26374630630016327 0.2470189332962036 36302192640
132
+ 2025-12-11 22:08:18 step 002 000014000 0.26374630630016327 0.2470189332962036 36302192640
133
+ 2025-12-11 22:08:20 best 002 000014000 0.26374630630016327 0.2470189332962036 36302192640
134
+ 2025-12-11 22:10:30 recent 003 000000000 0.24965786933898926 0.24705520272254944 36389928960
135
+ 2025-12-11 22:10:31 epoch 003 000000000 0.24965786933898926 0.24705520272254944 36389928960
136
+ 2025-12-11 22:25:04 recent 003 000001000 0.25500255823135376 0.2462674379348755 37250088960
137
+ 2025-12-11 22:25:06 step 003 000001000 0.25500255823135376 0.2462674379348755 37250088960
138
+ 2025-12-11 22:25:07 best 003 000001000 0.25500255823135376 0.2462674379348755 37250088960
139
+ 2025-12-11 22:39:39 recent 003 000002000 0.25009287148714066 0.24551482498645782 38110248960
140
+ 2025-12-11 22:39:41 step 003 000002000 0.25009287148714066 0.24551482498645782 38110248960
141
+ 2025-12-11 22:39:43 best 003 000002000 0.25009287148714066 0.24551482498645782 38110248960
142
+ 2025-12-11 22:54:18 recent 003 000003000 0.2639870420098305 0.2450481802225113 38970408960
143
+ 2025-12-11 22:54:20 step 003 000003000 0.2639870420098305 0.2450481802225113 38970408960
144
+ 2025-12-11 22:54:22 best 003 000003000 0.2639870420098305 0.2450481802225113 38970408960
145
+ 2025-12-11 23:08:56 recent 003 000004000 0.2620629668235779 0.24448950588703156 39830568960
146
+ 2025-12-11 23:08:58 step 003 000004000 0.2620629668235779 0.24448950588703156 39830568960
147
+ 2025-12-11 23:09:00 best 003 000004000 0.2620629668235779 0.24448950588703156 39830568960
148
+ 2025-12-11 23:23:35 recent 003 000005000 0.24179238826036453 0.24389499425888062 40690728960
149
+ 2025-12-11 23:23:37 step 003 000005000 0.24179238826036453 0.24389499425888062 40690728960
150
+ 2025-12-11 23:23:38 best 003 000005000 0.24179238826036453 0.24389499425888062 40690728960
151
+ 2025-12-11 23:38:07 recent 003 000006000 0.25250081717967987 0.2433004230260849 41550888960
152
+ 2025-12-11 23:38:09 step 003 000006000 0.25250081717967987 0.2433004230260849 41550888960
153
+ 2025-12-11 23:38:10 best 003 000006000 0.25250081717967987 0.2433004230260849 41550888960
154
+ 2025-12-11 23:52:46 recent 003 000007000 0.25653084367513657 0.2427988201379776 42411048960
155
+ 2025-12-11 23:52:47 step 003 000007000 0.25653084367513657 0.2427988201379776 42411048960
156
+ 2025-12-11 23:52:49 best 003 000007000 0.25653084367513657 0.2427988201379776 42411048960
157
+ 2025-12-12 00:07:25 recent 003 000008000 0.26132817566394806 0.24230463802814484 43271208960
158
+ 2025-12-12 00:07:27 step 003 000008000 0.26132817566394806 0.24230463802814484 43271208960
159
+ 2025-12-12 00:07:28 best 003 000008000 0.26132817566394806 0.24230463802814484 43271208960
160
+ 2025-12-12 00:22:03 recent 003 000009000 0.2560388073325157 0.24174869060516357 44131368960
161
+ 2025-12-12 00:22:05 step 003 000009000 0.2560388073325157 0.24174869060516357 44131368960
162
+ 2025-12-12 00:22:07 best 003 000009000 0.2560388073325157 0.24174869060516357 44131368960
163
+ 2025-12-12 00:36:41 recent 003 000010000 0.24657387286424637 0.24143913388252258 44991528960
164
+ 2025-12-12 00:36:43 step 003 000010000 0.24657387286424637 0.24143913388252258 44991528960
165
+ 2025-12-12 00:36:44 best 003 000010000 0.24657387286424637 0.24143913388252258 44991528960
166
+ 2025-12-12 00:51:16 recent 003 000011000 0.26763254404067993 0.24073179066181183 45851688960
167
+ 2025-12-12 00:51:17 step 003 000011000 0.26763254404067993 0.24073179066181183 45851688960
168
+ 2025-12-12 00:51:19 best 003 000011000 0.26763254404067993 0.24073179066181183 45851688960
169
+ 2025-12-12 01:05:54 recent 003 000012000 0.2579295039176941 0.24037586152553558 46711848960
170
+ 2025-12-12 01:05:56 step 003 000012000 0.2579295039176941 0.24037586152553558 46711848960
171
+ 2025-12-12 01:05:58 best 003 000012000 0.2579295039176941 0.24037586152553558 46711848960
172
+ 2025-12-12 01:20:33 recent 003 000013000 0.24819885194301605 0.23997525870800018 47572008960
173
+ 2025-12-12 01:20:35 step 003 000013000 0.24819885194301605 0.23997525870800018 47572008960
174
+ 2025-12-12 01:20:37 best 003 000013000 0.24819885194301605 0.23997525870800018 47572008960
175
+ 2025-12-12 01:35:10 recent 003 000014000 0.2604087218642235 0.23958879709243774 48432168960
176
+ 2025-12-12 01:35:12 step 003 000014000 0.2604087218642235 0.23958879709243774 48432168960
177
+ 2025-12-12 01:35:14 best 003 000014000 0.2604087218642235 0.23958879709243774 48432168960
178
+ 2025-12-12 01:37:23 recent 004 000000000 0.2494329959154129 0.23956745862960815 48519905280
179
+ 2025-12-12 01:37:25 epoch 004 000000000 0.2494329959154129 0.23956745862960815 48519905280
180
+ 2025-12-12 01:37:27 best 004 000000000 0.2494329959154129 0.23956745862960815 48519905280
181
+ 2025-12-12 01:52:01 recent 004 000001000 0.23722759634256363 0.2391868233680725 49380065280
182
+ 2025-12-12 01:52:02 step 004 000001000 0.23722759634256363 0.2391868233680725 49380065280
183
+ 2025-12-12 01:52:04 best 004 000001000 0.23722759634256363 0.2391868233680725 49380065280
184
+ 2025-12-12 02:06:35 recent 004 000002000 0.2478501945734024 0.23878975212574005 50240225280
185
+ 2025-12-12 02:06:37 step 004 000002000 0.2478501945734024 0.23878975212574005 50240225280
186
+ 2025-12-12 02:06:39 best 004 000002000 0.2478501945734024 0.23878975212574005 50240225280
187
+ 2025-12-12 02:21:13 recent 004 000003000 0.25706008821725845 0.23859435319900513 51100385280
188
+ 2025-12-12 02:21:14 step 004 000003000 0.25706008821725845 0.23859435319900513 51100385280
189
+ 2025-12-12 02:21:16 best 004 000003000 0.25706008821725845 0.23859435319900513 51100385280
190
+ 2025-12-12 02:35:50 recent 004 000004000 0.2420676127076149 0.2383161336183548 51960545280
191
+ 2025-12-12 02:35:52 step 004 000004000 0.2420676127076149 0.2383161336183548 51960545280
192
+ 2025-12-12 02:35:53 best 004 000004000 0.2420676127076149 0.2383161336183548 51960545280
193
+ 2025-12-12 02:50:28 recent 004 000005000 0.257579542696476 0.23810502886772156 52820705280
194
+ 2025-12-12 02:50:29 step 004 000005000 0.257579542696476 0.23810502886772156 52820705280
195
+ 2025-12-12 02:50:31 best 004 000005000 0.257579542696476 0.23810502886772156 52820705280
196
+ 2025-12-12 03:05:00 recent 004 000006000 0.23875396698713303 0.2378847599029541 53680865280
197
+ 2025-12-12 03:05:02 step 004 000006000 0.23875396698713303 0.2378847599029541 53680865280
198
+ 2025-12-12 03:05:03 best 004 000006000 0.23875396698713303 0.2378847599029541 53680865280
199
+ 2025-12-12 03:19:40 recent 004 000007000 0.2411988377571106 0.237748384475708 54541025280
200
+ 2025-12-12 03:19:41 step 004 000007000 0.2411988377571106 0.237748384475708 54541025280
201
+ 2025-12-12 03:19:43 best 004 000007000 0.2411988377571106 0.237748384475708 54541025280
202
+ 2025-12-12 03:34:18 recent 004 000008000 0.25034378468990326 0.2375868260860443 55401185280
203
+ 2025-12-12 03:34:19 step 004 000008000 0.25034378468990326 0.2375868260860443 55401185280
204
+ 2025-12-12 03:34:21 best 004 000008000 0.25034378468990326 0.2375868260860443 55401185280
205
+ 2025-12-12 03:48:57 recent 004 000009000 0.25600236654281616 0.23750117421150208 56261345280
206
+ 2025-12-12 03:48:59 step 004 000009000 0.25600236654281616 0.23750117421150208 56261345280
207
+ 2025-12-12 03:49:01 best 004 000009000 0.25600236654281616 0.23750117421150208 56261345280
208
+ 2025-12-12 04:03:32 recent 004 000010000 0.2378641739487648 0.23743970692157745 57121505280
209
+ 2025-12-12 04:03:34 step 004 000010000 0.2378641739487648 0.23743970692157745 57121505280
210
+ 2025-12-12 04:03:36 best 004 000010000 0.2378641739487648 0.23743970692157745 57121505280
211
+ 2025-12-12 04:18:07 recent 004 000011000 0.24954141676425934 0.2373601794242859 57981665280
212
+ 2025-12-12 04:18:09 step 004 000011000 0.24954141676425934 0.2373601794242859 57981665280
213
+ 2025-12-12 04:18:10 best 004 000011000 0.24954141676425934 0.2373601794242859 57981665280
214
+ 2025-12-12 04:32:46 recent 004 000012000 0.24604463577270508 0.23732398450374603 58841825280
215
+ 2025-12-12 04:32:48 step 004 000012000 0.24604463577270508 0.23732398450374603 58841825280
216
+ 2025-12-12 04:32:49 best 004 000012000 0.24604463577270508 0.23732398450374603 58841825280
217
+ 2025-12-12 04:47:25 recent 004 000013000 0.24328067898750305 0.23732741177082062 59701985280
218
+ 2025-12-12 04:47:27 step 004 000013000 0.24328067898750305 0.23732741177082062 59701985280
219
+ 2025-12-12 05:02:01 recent 004 000014000 0.23832398653030396 0.23732227087020874 60562145280
220
+ 2025-12-12 05:02:02 step 004 000014000 0.23832398653030396 0.23732227087020874 60562145280
221
+ 2025-12-12 05:02:04 best 004 000014000 0.23832398653030396 0.23732227087020874 60562145280
222
+ 2025-12-12 05:04:15 recent 005 000000000 0.2475239485502243 0.2373163253068924 60649881600
223
+ 2025-12-12 05:04:16 epoch 005 000000000 0.2475239485502243 0.2373163253068924 60649881600
224
+ 2025-12-12 05:04:18 best 005 000000000 0.2475239485502243 0.2373163253068924 60649881600
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000010000_tokens_57.1b_val=0.2374.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000011000_tokens_58b_val=0.2374.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000012000_tokens_58.8b_val=0.2373.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank1.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank2.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank3.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank4.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank5.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank6.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank7.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/gsm8k_results_epoch_004_step_000014000_tokens_60.6b_val=0.2373.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/transformer_resume/metrics_summary_epoch4_top5.csv CHANGED
@@ -1,2 +1,2 @@
1
  checkpoint,accuracy,correct,total
2
- epoch_004_step_000013000_tokens_59.7b_val=0.2434.pt,0.5671,748,1319
 
1
  checkpoint,accuracy,correct,total
2
+ epoch_004_step_000012000_tokens_58.8b_val=0.2373.pt,0.6096,804,1319
checkpoints/transformer_resume/verification_epoch_004_step_000010000_tokens_57.1b_val=0.2374.txt ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ ============================================================
2
+ FINAL MERGED RESULTS
3
+ ============================================================
4
+ Total: 824
5
+ Correct: 488
6
+ No Answer: 52
7
+ Accuracy: 59.22%
8
+ ============================================================
checkpoints/transformer_resume/verification_epoch_004_step_000011000_tokens_58b_val=0.2374.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 1319
2
+ Correct: 807
3
+ Incorrect: 512
4
+ No Answer: 68
5
+ Accuracy: 61.18%
checkpoints/transformer_resume/verification_epoch_004_step_000012000_tokens_58.8b_val=0.2373.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 1319
2
+ Correct: 804
3
+ Incorrect: 515
4
+ No Answer: 69
5
+ Accuracy: 60.96%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 1155
2
+ Correct: 697
3
+ Incorrect: 458
4
+ No Answer: 61
5
+ Accuracy: 60.35%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank1.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 106
3
+ Incorrect: 48
4
+ No Answer: 11
5
+ Accuracy: 64.24%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank2.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 113
3
+ Incorrect: 44
4
+ No Answer: 8
5
+ Accuracy: 68.48%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank3.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 101
3
+ Incorrect: 58
4
+ No Answer: 6
5
+ Accuracy: 61.21%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank4.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 96
3
+ Incorrect: 62
4
+ No Answer: 7
5
+ Accuracy: 58.18%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank5.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 98
3
+ Incorrect: 60
4
+ No Answer: 7
5
+ Accuracy: 59.39%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank6.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 165
2
+ Correct: 99
3
+ Incorrect: 55
4
+ No Answer: 11
5
+ Accuracy: 60.00%
checkpoints/transformer_resume/verification_epoch_004_step_000013000_tokens_59.7b_val=0.2373_rank7.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 164
2
+ Correct: 108
3
+ Incorrect: 51
4
+ No Answer: 5
5
+ Accuracy: 65.85%
checkpoints/transformer_resume/verification_epoch_004_step_000014000_tokens_60.6b_val=0.2373.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ Total: 1155
2
+ Correct: 731
3
+ Incorrect: 424
4
+ No Answer: 66
5
+ Accuracy: 63.29%