samoline commited on
Commit
8052592
·
verified ·
1 Parent(s): 3d8b045

Training in progress, step 2, checkpoint

Browse files
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:859e05a1cc5fe16916f012ee534aaeea294905d4cd297cf31d285efa3b4aff8d
3
  size 6764560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf6b8f8e6c1ae2c0979f92b9879d9ef98c1075de61c915d4d242d09a2cc4e700
3
  size 6764560
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a6314c847da605c81b9d1e9d0be1d14b618bbee4a29ec33ff8bbf0135a6c4d4
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:386b15c93893d130a9ec5b73b4223191040061565560b51ee899d46588f2337b
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:764cf599d0b83c5d302be0978d09e435a466a8511dc1f7e86d5ee3704818d46f
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2972d8cacb091bd0e98bfff54b2e4933301753087ce149a9bf9a205a5e2d795
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 2.867712425797941e-05,
5
  "eval_steps": 1,
6
- "global_step": 1,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -22,6 +22,21 @@
22
  "eval_samples_per_second": 15.565,
23
  "eval_steps_per_second": 15.565,
24
  "step": 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  }
26
  ],
27
  "logging_steps": 1,
@@ -36,12 +51,12 @@
36
  "should_evaluate": false,
37
  "should_log": false,
38
  "should_save": true,
39
- "should_training_stop": false
40
  },
41
  "attributes": {}
42
  }
43
  },
44
- "total_flos": 660016594944.0,
45
  "train_batch_size": 1,
46
  "trial_name": null,
47
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 5.735424851595882e-05,
5
  "eval_steps": 1,
6
+ "global_step": 2,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
22
  "eval_samples_per_second": 15.565,
23
  "eval_steps_per_second": 15.565,
24
  "step": 1
25
+ },
26
+ {
27
+ "epoch": 5.735424851595882e-05,
28
+ "grad_norm": NaN,
29
+ "learning_rate": 4e-05,
30
+ "loss": 0.0,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 5.735424851595882e-05,
35
+ "eval_loss": NaN,
36
+ "eval_runtime": 114.8171,
37
+ "eval_samples_per_second": 15.991,
38
+ "eval_steps_per_second": 15.991,
39
+ "step": 2
40
  }
41
  ],
42
  "logging_steps": 1,
 
51
  "should_evaluate": false,
52
  "should_log": false,
53
  "should_save": true,
54
+ "should_training_stop": true
55
  },
56
  "attributes": {}
57
  }
58
  },
59
+ "total_flos": 990024892416.0,
60
  "train_batch_size": 1,
61
  "trial_name": null,
62
  "trial_params": null