duyvt6663 commited on
Commit
85d64f7
·
1 Parent(s): 02887bf

Training in progress, step 200, checkpoint

Browse files
checkpoint-200/README.md CHANGED
@@ -216,23 +216,4 @@ The following `bitsandbytes` quantization config was used during training:
216
  ### Framework versions
217
 
218
 
219
- - PEFT 0.6.0.dev0
220
- ## Training procedure
221
-
222
-
223
- The following `bitsandbytes` quantization config was used during training:
224
- - quant_method: bitsandbytes
225
- - load_in_8bit: True
226
- - load_in_4bit: False
227
- - llm_int8_threshold: 6.0
228
- - llm_int8_skip_modules: None
229
- - llm_int8_enable_fp32_cpu_offload: False
230
- - llm_int8_has_fp16_weight: False
231
- - bnb_4bit_quant_type: fp4
232
- - bnb_4bit_use_double_quant: False
233
- - bnb_4bit_compute_dtype: float32
234
-
235
- ### Framework versions
236
-
237
-
238
  - PEFT 0.6.0.dev0
 
216
  ### Framework versions
217
 
218
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
219
  - PEFT 0.6.0.dev0
checkpoint-200/adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4e4650c6715b7603246bfd038984d45d9685d9f7a07ba6a9f234044e77c1c76e
3
  size 9873829
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c915b534dd5c63af4a4da8b51c402aa67519f4f509b5abe822440b79d0dada7
3
  size 9873829
checkpoint-200/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72618a45d8a50c4f5e08ee8a9c5515001f65ebc6b695979e3054665f22a31d46
3
  size 42724
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e98623969976c0a14b109a3c67209e22c6ee22bd13f81248dbbc28c82287acfc
3
  size 42724
checkpoint-200/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b7d888b148eb21b791090fe08933aa4e2a93c346d8904e0237d31802a4c179b
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24ca455dde448b70aa653e0498d3346b98980a6d89972d5d56f2b2ec8d9c7be4
3
  size 14244
checkpoint-200/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f2c7ad61cb507f1edc7b5922a62a9509826b96a88d5fa2c74a8bf3946bfd30f
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d191efbbc9b0763e805de64fb2ed1cd1a38712d135eda6d145caf5703e52c386
3
  size 1064
checkpoint-200/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.665068507194519,
3
- "best_model_checkpoint": "output/checkpoint-50",
4
- "epoch": 0.8184143222506394,
5
  "eval_steps": 50,
6
  "global_step": 200,
7
  "is_hyper_param_search": false,
@@ -9,77 +9,77 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.0,
13
- "learning_rate": 7.575757575757576e-07,
14
- "loss": 0.4968,
15
  "step": 1
16
  },
17
  {
18
- "epoch": 0.2,
19
- "learning_rate": 3.484848484848485e-05,
20
- "loss": 0.5254,
21
  "step": 50
22
  },
23
  {
24
- "epoch": 0.2,
25
- "eval_accuracy": 0.6507462686567164,
26
- "eval_loss": 0.665068507194519,
27
- "eval_runtime": 61.5592,
28
- "eval_samples_per_second": 5.442,
29
- "eval_steps_per_second": 1.365,
30
  "step": 50
31
  },
32
  {
33
- "epoch": 0.41,
34
- "learning_rate": 7.272727272727273e-05,
35
- "loss": 0.4971,
36
  "step": 100
37
  },
38
  {
39
- "epoch": 0.41,
40
- "eval_accuracy": 0.6656716417910448,
41
- "eval_loss": 0.8002150058746338,
42
- "eval_runtime": 61.211,
43
- "eval_samples_per_second": 5.473,
44
- "eval_steps_per_second": 1.372,
45
  "step": 100
46
  },
47
  {
48
- "epoch": 0.61,
49
- "learning_rate": 9.783333333333334e-05,
50
- "loss": 0.5039,
51
  "step": 150
52
  },
53
  {
54
- "epoch": 0.61,
55
- "eval_accuracy": 0.6865671641791045,
56
- "eval_loss": 0.7404947280883789,
57
- "eval_runtime": 61.4487,
58
- "eval_samples_per_second": 5.452,
59
- "eval_steps_per_second": 1.367,
60
  "step": 150
61
  },
62
  {
63
- "epoch": 0.82,
64
- "learning_rate": 8.950000000000001e-05,
65
- "loss": 0.4944,
66
  "step": 200
67
  },
68
  {
69
- "epoch": 0.82,
70
- "eval_accuracy": 0.6238805970149254,
71
- "eval_loss": 0.9986834526062012,
72
- "eval_runtime": 60.5135,
73
- "eval_samples_per_second": 5.536,
74
- "eval_steps_per_second": 1.388,
75
  "step": 200
76
  }
77
  ],
78
  "logging_steps": 50,
79
- "max_steps": 732,
80
- "num_train_epochs": 3,
81
  "save_steps": 50,
82
- "total_flos": 8.514407519772672e+16,
83
  "trial_name": null,
84
  "trial_params": null
85
  }
 
1
  {
2
+ "best_metric": 0.5725014805793762,
3
+ "best_model_checkpoint": "output/checkpoint-100",
4
+ "epoch": 3.7735849056603774,
5
  "eval_steps": 50,
6
  "global_step": 200,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.02,
13
+ "learning_rate": 1.234567901234568e-06,
14
+ "loss": 0.4437,
15
  "step": 1
16
  },
17
  {
18
+ "epoch": 0.94,
19
+ "learning_rate": 5.679012345679012e-05,
20
+ "loss": 0.455,
21
  "step": 50
22
  },
23
  {
24
+ "epoch": 0.94,
25
+ "eval_accuracy": 0.68,
26
+ "eval_loss": 0.6171127557754517,
27
+ "eval_runtime": 29.0494,
28
+ "eval_samples_per_second": 6.024,
29
+ "eval_steps_per_second": 1.515,
30
  "step": 50
31
  },
32
  {
33
+ "epoch": 1.89,
34
+ "learning_rate": 9.59349593495935e-05,
35
+ "loss": 0.4555,
36
  "step": 100
37
  },
38
  {
39
+ "epoch": 1.89,
40
+ "eval_accuracy": 0.7257142857142858,
41
+ "eval_loss": 0.5725014805793762,
42
+ "eval_runtime": 28.7986,
43
+ "eval_samples_per_second": 6.077,
44
+ "eval_steps_per_second": 1.528,
45
  "step": 100
46
  },
47
  {
48
+ "epoch": 2.83,
49
+ "learning_rate": 8.265582655826559e-05,
50
+ "loss": 0.455,
51
  "step": 150
52
  },
53
  {
54
+ "epoch": 2.83,
55
+ "eval_accuracy": 0.7371428571428571,
56
+ "eval_loss": 0.5729417204856873,
57
+ "eval_runtime": 28.9054,
58
+ "eval_samples_per_second": 6.054,
59
+ "eval_steps_per_second": 1.522,
60
  "step": 150
61
  },
62
  {
63
+ "epoch": 3.77,
64
+ "learning_rate": 6.910569105691057e-05,
65
+ "loss": 0.4718,
66
  "step": 200
67
  },
68
  {
69
+ "epoch": 3.77,
70
+ "eval_accuracy": 0.6685714285714286,
71
+ "eval_loss": 0.6322054862976074,
72
+ "eval_runtime": 28.8878,
73
+ "eval_samples_per_second": 6.058,
74
+ "eval_steps_per_second": 1.523,
75
  "step": 200
76
  }
77
  ],
78
  "logging_steps": 50,
79
+ "max_steps": 450,
80
+ "num_train_epochs": 9,
81
  "save_steps": 50,
82
+ "total_flos": 7.652955795308544e+16,
83
  "trial_name": null,
84
  "trial_params": null
85
  }
checkpoint-200/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9feea516a8a208c393c324d84d047c0ab5036bbdf3fb1f3eea6a2563e9c79a6b
3
  size 4472
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91c1e3ec7fc28d5ed2a9d783b7ded0a7affbbd029319e299d9244ac0f3fbc29
3
  size 4472