tmnam20 commited on
Commit
35b1422
1 Parent(s): c7dafcc

Training in progress, step 3000, checkpoint

Browse files
checkpoint-3000/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9bd8b8d005c8498d0dc87f4cc00ffebdf7a4f1c4fbaf9c06374d9b4bb13b658f
3
  size 932401285
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de7d6f6157cff9810cf8ea9f033c4fd681347531ec68cf04dea329e43b0c4db8
3
  size 932401285
checkpoint-3000/pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c1404200e7060273f92a189004bc61a4f6d9ed1be64b6433f51dc931e1da4576
3
  size 466205145
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0e01d89e2c093d6586b64996b2d730467c6eec8ba6acb4fbf7c776362f563fe
3
  size 466205145
checkpoint-3000/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:73a860bf7adbc7134c78671fd5c6e6588c8b56d0bda60913c03b8c144c3849d7
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7249c42f48d99a7463b035eeb788a527848411ab95b413e6f1032df9ad08c927
3
  size 14575
checkpoint-3000/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 1.1441665887832642,
3
  "best_model_checkpoint": "/kaggle/output/checkpoint-3000",
4
- "epoch": 48.0,
5
  "eval_steps": 1000,
6
  "global_step": 3000,
7
  "is_hyper_param_search": false,
@@ -9,60 +9,60 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.02,
13
  "learning_rate": 2.7777777777777777e-11,
14
- "loss": 1.2395,
15
  "step": 1
16
  },
17
  {
18
- "epoch": 16.0,
19
  "learning_rate": 2.7638888888888893e-08,
20
- "loss": 1.1422,
21
  "step": 1000
22
  },
23
  {
24
- "epoch": 16.0,
25
- "eval_accuracy": 0.33073852295409184,
26
- "eval_loss": 1.1557353734970093,
27
- "eval_runtime": 54.1805,
28
- "eval_samples_per_second": 92.469,
29
- "eval_steps_per_second": 11.572,
30
  "step": 1000
31
  },
32
  {
33
- "epoch": 32.0,
34
  "learning_rate": 5.541666666666667e-08,
35
- "loss": 1.1401,
36
  "step": 2000
37
  },
38
  {
39
- "epoch": 32.0,
40
- "eval_accuracy": 0.3311377245508982,
41
- "eval_loss": 1.1507551670074463,
42
- "eval_runtime": 53.98,
43
- "eval_samples_per_second": 92.812,
44
- "eval_steps_per_second": 11.615,
45
  "step": 2000
46
  },
47
  {
48
- "epoch": 48.0,
49
  "learning_rate": 8.316666666666666e-08,
50
- "loss": 1.1287,
51
  "step": 3000
52
  },
53
  {
54
- "epoch": 48.0,
55
- "eval_accuracy": 0.3321357285429142,
56
- "eval_loss": 1.1441665887832642,
57
- "eval_runtime": 53.9873,
58
- "eval_samples_per_second": 92.8,
59
- "eval_steps_per_second": 11.614,
60
  "step": 3000
61
  }
62
  ],
63
  "logging_steps": 1000,
64
  "max_steps": 10000000,
65
- "num_train_epochs": 161291,
66
  "save_steps": 1000,
67
  "total_flos": 6271065391104000.0,
68
  "trial_name": null,
 
1
  {
2
+ "best_metric": 1.127414345741272,
3
  "best_model_checkpoint": "/kaggle/output/checkpoint-3000",
4
+ "epoch": 0.12222946544980444,
5
  "eval_steps": 1000,
6
  "global_step": 3000,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.0,
13
  "learning_rate": 2.7777777777777777e-11,
14
+ "loss": 1.029,
15
  "step": 1
16
  },
17
  {
18
+ "epoch": 0.04,
19
  "learning_rate": 2.7638888888888893e-08,
20
+ "loss": 1.181,
21
  "step": 1000
22
  },
23
  {
24
+ "epoch": 0.04,
25
+ "eval_accuracy": 0.3273453093812375,
26
+ "eval_loss": 1.1529844999313354,
27
+ "eval_runtime": 54.2837,
28
+ "eval_samples_per_second": 92.293,
29
+ "eval_steps_per_second": 11.55,
30
  "step": 1000
31
  },
32
  {
33
+ "epoch": 0.08,
34
  "learning_rate": 5.541666666666667e-08,
35
+ "loss": 1.1527,
36
  "step": 2000
37
  },
38
  {
39
+ "epoch": 0.08,
40
+ "eval_accuracy": 0.33013972055888224,
41
+ "eval_loss": 1.1351025104522705,
42
+ "eval_runtime": 54.0518,
43
+ "eval_samples_per_second": 92.689,
44
+ "eval_steps_per_second": 11.6,
45
  "step": 2000
46
  },
47
  {
48
+ "epoch": 0.12,
49
  "learning_rate": 8.316666666666666e-08,
50
+ "loss": 1.142,
51
  "step": 3000
52
  },
53
  {
54
+ "epoch": 0.12,
55
+ "eval_accuracy": 0.3317365269461078,
56
+ "eval_loss": 1.127414345741272,
57
+ "eval_runtime": 54.0871,
58
+ "eval_samples_per_second": 92.628,
59
+ "eval_steps_per_second": 11.592,
60
  "step": 3000
61
  }
62
  ],
63
  "logging_steps": 1000,
64
  "max_steps": 10000000,
65
+ "num_train_epochs": 408,
66
  "save_steps": 1000,
67
  "total_flos": 6271065391104000.0,
68
  "trial_name": null,
checkpoint-3000/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cb2cd1b7e648a386ccb79e47a451842e11001981c4363e993b678e0818db5cab
3
  size 4091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5520446b18da09f4a43173b75839457981a344c0e709fd4bbcb249e3f60c859e
3
  size 4091