update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: distilgpt2
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
model-index:
|
7 |
+
- name: distilgpt2-sd
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# distilgpt2-sd
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 1.4481
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
More information needed
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Training and evaluation data
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training procedure
|
33 |
+
|
34 |
+
### Training hyperparameters
|
35 |
+
|
36 |
+
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 5e-05
|
38 |
+
- train_batch_size: 8
|
39 |
+
- eval_batch_size: 8
|
40 |
+
- seed: 42
|
41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
+
- lr_scheduler_type: linear
|
43 |
+
- num_epochs: 3
|
44 |
+
|
45 |
+
### Training results
|
46 |
+
|
47 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
48 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
49 |
+
| 3.36 | 0.05 | 500 | 2.8209 |
|
50 |
+
| 2.8086 | 0.11 | 1000 | 2.5757 |
|
51 |
+
| 2.6126 | 0.16 | 1500 | 2.4096 |
|
52 |
+
| 2.4771 | 0.22 | 2000 | 2.3027 |
|
53 |
+
| 2.3986 | 0.27 | 2500 | 2.2076 |
|
54 |
+
| 2.3148 | 0.33 | 3000 | 2.1547 |
|
55 |
+
| 2.237 | 0.38 | 3500 | 2.0825 |
|
56 |
+
| 2.1731 | 0.43 | 4000 | 2.0334 |
|
57 |
+
| 2.1256 | 0.49 | 4500 | 1.9806 |
|
58 |
+
| 2.081 | 0.54 | 5000 | 1.9345 |
|
59 |
+
| 2.0677 | 0.6 | 5500 | 1.9053 |
|
60 |
+
| 1.9794 | 0.65 | 6000 | 1.8691 |
|
61 |
+
| 2.0072 | 0.71 | 6500 | 1.8429 |
|
62 |
+
| 1.9597 | 0.76 | 7000 | 1.8061 |
|
63 |
+
| 1.9318 | 0.82 | 7500 | 1.7857 |
|
64 |
+
| 1.9283 | 0.87 | 8000 | 1.7610 |
|
65 |
+
| 1.8959 | 0.92 | 8500 | 1.7378 |
|
66 |
+
| 1.8626 | 0.98 | 9000 | 1.7185 |
|
67 |
+
| 1.8126 | 1.03 | 9500 | 1.7040 |
|
68 |
+
| 1.7789 | 1.09 | 10000 | 1.6855 |
|
69 |
+
| 1.7794 | 1.14 | 10500 | 1.6756 |
|
70 |
+
| 1.7284 | 1.2 | 11000 | 1.6529 |
|
71 |
+
| 1.7478 | 1.25 | 11500 | 1.6384 |
|
72 |
+
| 1.7065 | 1.3 | 12000 | 1.6321 |
|
73 |
+
| 1.7092 | 1.36 | 12500 | 1.6133 |
|
74 |
+
| 1.6897 | 1.41 | 13000 | 1.6146 |
|
75 |
+
| 1.6902 | 1.47 | 13500 | 1.5952 |
|
76 |
+
| 1.6888 | 1.52 | 14000 | 1.5792 |
|
77 |
+
| 1.6862 | 1.58 | 14500 | 1.5730 |
|
78 |
+
| 1.6458 | 1.63 | 15000 | 1.5661 |
|
79 |
+
| 1.6594 | 1.68 | 15500 | 1.5537 |
|
80 |
+
| 1.6486 | 1.74 | 16000 | 1.5484 |
|
81 |
+
| 1.6556 | 1.79 | 16500 | 1.5360 |
|
82 |
+
| 1.6187 | 1.85 | 17000 | 1.5264 |
|
83 |
+
| 1.6377 | 1.9 | 17500 | 1.5223 |
|
84 |
+
| 1.6129 | 1.96 | 18000 | 1.5180 |
|
85 |
+
| 1.6025 | 2.01 | 18500 | 1.5030 |
|
86 |
+
| 1.5697 | 2.06 | 19000 | 1.4991 |
|
87 |
+
| 1.5616 | 2.12 | 19500 | 1.5012 |
|
88 |
+
| 1.558 | 2.17 | 20000 | 1.4984 |
|
89 |
+
| 1.549 | 2.23 | 20500 | 1.4809 |
|
90 |
+
| 1.5048 | 2.28 | 21000 | 1.4827 |
|
91 |
+
| 1.5207 | 2.34 | 21500 | 1.4740 |
|
92 |
+
| 1.5097 | 2.39 | 22000 | 1.4699 |
|
93 |
+
| 1.541 | 2.45 | 22500 | 1.4701 |
|
94 |
+
| 1.5355 | 2.5 | 23000 | 1.4637 |
|
95 |
+
| 1.5318 | 2.55 | 23500 | 1.4609 |
|
96 |
+
| 1.5352 | 2.61 | 24000 | 1.4580 |
|
97 |
+
| 1.5202 | 2.66 | 24500 | 1.4566 |
|
98 |
+
| 1.5073 | 2.72 | 25000 | 1.4547 |
|
99 |
+
| 1.5462 | 2.77 | 25500 | 1.4520 |
|
100 |
+
| 1.5347 | 2.83 | 26000 | 1.4491 |
|
101 |
+
| 1.52 | 2.88 | 26500 | 1.4488 |
|
102 |
+
| 1.5154 | 2.93 | 27000 | 1.4475 |
|
103 |
+
| 1.4855 | 2.99 | 27500 | 1.4481 |
|
104 |
+
|
105 |
+
|
106 |
+
### Framework versions
|
107 |
+
|
108 |
+
- Transformers 4.31.0
|
109 |
+
- Pytorch 2.0.1+cu118
|
110 |
+
- Datasets 2.13.1
|
111 |
+
- Tokenizers 0.13.3
|