Hugging Face's logo

beomi
/
KoAlpaca-Polyglot-5.8B

Text Generation
Transformers
PyTorch
Safetensors
Korean
gpt_neox
Generated from Trainer
polyglot-ko
gpt-neox
KoAlpaca
Inference Endpoints
text-generation-inference
Model card Files Community
2
june42 commited on
Commit
b7438fb
•
1 Parent(s): 4a2dfd1

Finetune model with following https://github.com/tatsu-lab/stanford_alpaca#fine-tuning

Browse files
Files changed (7) hide show
  1. optimizer.pt +1 -1
  2. pytorch_model-00001-of-00002.bin +1 -1
  3. pytorch_model-00002-of-00002.bin +1 -1
  4. rng_state.pth +1 -1
  5. scheduler.pt +1 -1
  6. trainer_state.json +43 -15
  7. training_args.bin +1 -1
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1e019f0487293f7bdede232d061e4d39315100768ca7a66b14318586e487b623
3
  size 13835141
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a06652ab49c35704c97135a68936d8724b53751af0a922d0df4bf036f272a79f
3
  size 13835141
pytorch_model-00001-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f559f3e37d5fe208f0841dc756dbf80b149d19a6dc173ab683026def9530d3f
3
  size 9979879204
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d31b0599bec571926fcbb4c8d74df6a54b9f856658d578e0bfc91f50f81ef2ef
3
  size 9979879204
pytorch_model-00002-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:81f5ea8e8276cf51325f479e1d9e83c44f1022ad5f3b66fac903b9f336c72509
3
  size 2025273825
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8577b0c87cbe8310a1d2cc2f66be9a972f6c003663538dcb7d57aad5f0353c8
3
  size 2025273825
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dcd12d19c79af278f28d08dd9e2cd392161951f2f9fa3355e120be3eeb3cee42
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cde37b1d59ec8275aa116b67c6503ca997bd9a90886b106af365cbb8dfef4db9
3
  size 14575
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:61d142a00f56e370dd08ab7f419d99eb71c744e7cf0ae126b6069a35d7c59ca7
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e7bdefe584e61a57a129ab2768e96850498503bfc68609b940b74cde526cebd
3
  size 627
trainer_state.json CHANGED
@@ -1,36 +1,64 @@
1
  {
2
- "best_metric": 0.252197265625,
3
- "best_model_checkpoint": "./models/2023-03-15_1edc3387-3214/checkpoint-5067",
4
- "epoch": 1.0,
5
- "global_step": 5067,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.0,
12
- "learning_rate": 1.999868429708572e-05,
13
- "loss": 10.1719,
14
  "step": 1
15
  },
16
  {
17
  "epoch": 1.0,
18
  "learning_rate": 1.3333333333333333e-05,
19
- "loss": 0.2619,
20
- "step": 5067
21
  },
22
  {
23
  "epoch": 1.0,
24
- "eval_loss": 0.252197265625,
25
- "eval_runtime": 325.5581,
26
- "eval_samples_per_second": 13.835,
27
- "eval_steps_per_second": 1.729,
28
- "step": 5067
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  }
30
  ],
31
- "max_steps": 15201,
32
  "num_train_epochs": 3,
33
- "total_flos": 7.17503717456216e+17,
34
  "trial_name": null,
35
  "trial_params": null
36
  }
 
1
  {
2
+ "best_metric": 0.2509765625,
3
+ "best_model_checkpoint": "./models/2023-03-16_1edc4008-e54e/checkpoint-2534",
4
+ "epoch": 3.0,
5
+ "global_step": 7602,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.0,
12
+ "learning_rate": 1.9997369113391214e-05,
13
+ "loss": 9.8984,
14
  "step": 1
15
  },
16
  {
17
  "epoch": 1.0,
18
  "learning_rate": 1.3333333333333333e-05,
19
+ "loss": 0.2605,
20
+ "step": 2534
21
  },
22
  {
23
  "epoch": 1.0,
24
+ "eval_loss": 0.2509765625,
25
+ "eval_runtime": 191.3685,
26
+ "eval_samples_per_second": 23.536,
27
+ "eval_steps_per_second": 2.942,
28
+ "step": 2534
29
+ },
30
+ {
31
+ "epoch": 2.0,
32
+ "learning_rate": 6.666666666666667e-06,
33
+ "loss": 0.1598,
34
+ "step": 5068
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_loss": 0.258544921875,
39
+ "eval_runtime": 159.499,
40
+ "eval_samples_per_second": 28.238,
41
+ "eval_steps_per_second": 3.53,
42
+ "step": 5068
43
+ },
44
+ {
45
+ "epoch": 3.0,
46
+ "learning_rate": 0.0,
47
+ "loss": 0.0753,
48
+ "step": 7602
49
+ },
50
+ {
51
+ "epoch": 3.0,
52
+ "eval_loss": 0.3056640625,
53
+ "eval_runtime": 159.3863,
54
+ "eval_samples_per_second": 28.258,
55
+ "eval_steps_per_second": 3.532,
56
+ "step": 7602
57
  }
58
  ],
59
+ "max_steps": 7602,
60
  "num_train_epochs": 3,
61
+ "total_flos": 2.1525111523686482e+18,
62
  "trial_name": null,
63
  "trial_params": null
64
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:06fdea69e1b755dd45733f4f1ca4d1dcabbbee294889226ecb3c1020cb98963d
3
  size 3323
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4872b8314368e34757a8c3791ebfcf77a89ff4d439e81f25e5640f42c35bec74
3
  size 3323