thorirhrafn commited on
Commit
e4a8444
·
verified ·
1 Parent(s): 038b44b

End of training

Browse files
Files changed (1) hide show
  1. README.md +88 -0
README.md ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - reward-trainer
7
+ - generated_from_trainer
8
+ metrics:
9
+ - accuracy
10
+ base_model: AI-Sweden-Models/gpt-sw3-1.3b
11
+ model-index:
12
+ - name: gpt1B_reward_test
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # gpt1B_reward_test
20
+
21
+ This model is a fine-tuned version of [AI-Sweden-Models/gpt-sw3-1.3b](https://huggingface.co/AI-Sweden-Models/gpt-sw3-1.3b) on an unknown dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.6367
24
+ - Accuracy: 0.6504
25
+
26
+ ## Model description
27
+
28
+ More information needed
29
+
30
+ ## Intended uses & limitations
31
+
32
+ More information needed
33
+
34
+ ## Training and evaluation data
35
+
36
+ More information needed
37
+
38
+ ## Training procedure
39
+
40
+ ### Training hyperparameters
41
+
42
+ The following hyperparameters were used during training:
43
+ - learning_rate: 3e-05
44
+ - train_batch_size: 2
45
+ - eval_batch_size: 2
46
+ - seed: 42
47
+ - gradient_accumulation_steps: 8
48
+ - total_train_batch_size: 16
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - num_epochs: 1
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
56
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
57
+ | 0.6878 | 0.04 | 200 | 0.6850 | 0.5791 |
58
+ | 0.6724 | 0.08 | 400 | 0.6740 | 0.6024 |
59
+ | 0.6611 | 0.13 | 600 | 0.6703 | 0.6081 |
60
+ | 0.6435 | 0.17 | 800 | 0.6773 | 0.6036 |
61
+ | 0.6787 | 0.21 | 1000 | 0.6544 | 0.6189 |
62
+ | 0.7166 | 0.25 | 1200 | 0.6697 | 0.6223 |
63
+ | 0.614 | 0.3 | 1400 | 0.6590 | 0.6250 |
64
+ | 0.6279 | 0.34 | 1600 | 0.6422 | 0.6343 |
65
+ | 0.6185 | 0.38 | 1800 | 0.6427 | 0.6389 |
66
+ | 0.5539 | 0.42 | 2000 | 0.6459 | 0.6390 |
67
+ | 0.5988 | 0.47 | 2200 | 0.6485 | 0.6379 |
68
+ | 0.6096 | 0.51 | 2400 | 0.6570 | 0.6439 |
69
+ | 0.5898 | 0.55 | 2600 | 0.6381 | 0.6441 |
70
+ | 0.6366 | 0.59 | 2800 | 0.6479 | 0.6389 |
71
+ | 0.6457 | 0.64 | 3000 | 0.6397 | 0.6490 |
72
+ | 0.6171 | 0.68 | 3200 | 0.6476 | 0.6467 |
73
+ | 0.5262 | 0.72 | 3400 | 0.6506 | 0.6458 |
74
+ | 0.5723 | 0.76 | 3600 | 0.6467 | 0.6471 |
75
+ | 0.6194 | 0.81 | 3800 | 0.6393 | 0.6480 |
76
+ | 0.5946 | 0.85 | 4000 | 0.6375 | 0.6490 |
77
+ | 0.5868 | 0.89 | 4200 | 0.6366 | 0.6503 |
78
+ | 0.5905 | 0.93 | 4400 | 0.6367 | 0.6505 |
79
+ | 0.5651 | 0.97 | 4600 | 0.6367 | 0.6504 |
80
+
81
+
82
+ ### Framework versions
83
+
84
+ - PEFT 0.8.2
85
+ - Transformers 4.38.1
86
+ - Pytorch 2.2.0+cu118
87
+ - Datasets 2.17.1
88
+ - Tokenizers 0.15.2