flytech commited on
Commit
7693fa1
·
1 Parent(s): 1125538

Training in progress, step 300, checkpoint

Browse files
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c2f0d7b34ba8adad3475e88cd4a5bc160d8ee2ec2c7661852d779c0976d7191b
3
  size 159967880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4210658d26eeba1aa3e43cb8b568650858990813e984155136bb1252c70a7f7
3
  size 159967880
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:38d74a1a36db14db8e6ed6d04edd1fb572ce2feac9fec5b1a6d5f84aa066bb7d
3
- size 80630175
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6ef2dbeed00cdc9f5a6258f103ac9d74aa429a1686555384e0c3a8c71ceec09
3
+ size 80630623
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9735754b23d5a4ee87fabc8c3b06ba8e506c17129863b83f1f5e6e43f52cee3
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13a1f8535e17826b08dca69f1587ba6a678d164c9d6edd72af2ee4c24f50b33f
3
  size 14575
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1bca9d737fd20d20ec53fe917bba1c7d096ffa7b89b6ad2bdf6b7d5b1b989cd5
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9218b00f81cdc7c95a459552cbf25b1fbc52bb38d9c85044830d891ce9d12f83
3
  size 627
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.2195121951219512,
5
  "eval_steps": 100,
6
- "global_step": 200,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -33,13 +33,26 @@
33
  "eval_samples_per_second": 1.858,
34
  "eval_steps_per_second": 0.242,
35
  "step": 200
 
 
 
 
 
 
 
 
 
 
 
 
 
36
  }
37
  ],
38
  "logging_steps": 100,
39
  "max_steps": 1804,
40
  "num_train_epochs": 11,
41
  "save_steps": 100,
42
- "total_flos": 1.6726059009441792e+16,
43
  "trial_name": null,
44
  "trial_params": null
45
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 1.8292682926829267,
5
  "eval_steps": 100,
6
+ "global_step": 300,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
33
  "eval_samples_per_second": 1.858,
34
  "eval_steps_per_second": 0.242,
35
  "step": 200
36
+ },
37
+ {
38
+ "epoch": 1.83,
39
+ "learning_rate": 0.0002,
40
+ "loss": 0.1129,
41
+ "step": 300
42
+ },
43
+ {
44
+ "epoch": 1.83,
45
+ "eval_runtime": 37.2493,
46
+ "eval_samples_per_second": 1.852,
47
+ "eval_steps_per_second": 0.242,
48
+ "step": 300
49
  }
50
  ],
51
  "logging_steps": 100,
52
  "max_steps": 1804,
53
  "num_train_epochs": 11,
54
  "save_steps": 100,
55
+ "total_flos": 2.510479865099059e+16,
56
  "trial_name": null,
57
  "trial_params": null
58
  }