Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
model-index:
|
6 |
+
- name: EnabledChat
|
7 |
+
results: []
|
8 |
+
pipeline_tag: text-generation
|
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 |
+
# EnabledChat
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [facebook/opt-1.3b](https://huggingface.co/facebook/opt-1.3b) on the None dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 1.1152
|
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: 0.0002
|
38 |
+
- train_batch_size: 16
|
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 |
+
- lr_scheduler_warmup_steps: 100
|
44 |
+
- num_epochs: 3
|
45 |
+
- mixed_precision_training: Native AMP
|
46 |
+
|
47 |
+
### Training results
|
48 |
+
|
49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
51 |
+
| 1.2849 | 0.06 | 200 | 1.2346 |
|
52 |
+
| 1.2515 | 0.13 | 400 | 1.2077 |
|
53 |
+
| 1.1859 | 0.19 | 600 | 1.1948 |
|
54 |
+
| 1.2007 | 0.26 | 800 | 1.1845 |
|
55 |
+
| 1.2085 | 0.32 | 1000 | 1.1763 |
|
56 |
+
| 1.2036 | 0.38 | 1200 | 1.1710 |
|
57 |
+
| 1.2134 | 0.45 | 1400 | 1.1655 |
|
58 |
+
| 1.1869 | 0.51 | 1600 | 1.1610 |
|
59 |
+
| 1.2017 | 0.58 | 1800 | 1.1573 |
|
60 |
+
| 1.2139 | 0.64 | 2000 | 1.1537 |
|
61 |
+
| 1.1302 | 0.7 | 2200 | 1.1510 |
|
62 |
+
| 1.1799 | 0.77 | 2400 | 1.1480 |
|
63 |
+
| 1.1518 | 0.83 | 2600 | 1.1461 |
|
64 |
+
| 1.1752 | 0.9 | 2800 | 1.1434 |
|
65 |
+
| 1.1584 | 0.96 | 3000 | 1.1415 |
|
66 |
+
| 1.1311 | 1.02 | 3200 | 1.1400 |
|
67 |
+
| 1.1595 | 1.09 | 3400 | 1.1374 |
|
68 |
+
| 1.1721 | 1.15 | 3600 | 1.1365 |
|
69 |
+
| 1.1597 | 1.22 | 3800 | 1.1355 |
|
70 |
+
| 1.1718 | 1.28 | 4000 | 1.1335 |
|
71 |
+
| 1.1639 | 1.34 | 4200 | 1.1315 |
|
72 |
+
| 1.1448 | 1.41 | 4400 | 1.1312 |
|
73 |
+
| 1.1645 | 1.47 | 4600 | 1.1296 |
|
74 |
+
| 1.1562 | 1.54 | 4800 | 1.1282 |
|
75 |
+
| 1.1619 | 1.6 | 5000 | 1.1273 |
|
76 |
+
| 1.1794 | 1.66 | 5200 | 1.1270 |
|
77 |
+
| 1.178 | 1.73 | 5400 | 1.1254 |
|
78 |
+
| 1.1291 | 1.79 | 5600 | 1.1246 |
|
79 |
+
| 1.1235 | 1.86 | 5800 | 1.1234 |
|
80 |
+
| 1.1169 | 1.92 | 6000 | 1.1222 |
|
81 |
+
| 1.1494 | 1.98 | 6200 | 1.1214 |
|
82 |
+
| 1.1032 | 2.05 | 6400 | 1.1218 |
|
83 |
+
| 1.1269 | 2.11 | 6600 | 1.1207 |
|
84 |
+
| 1.1561 | 2.18 | 6800 | 1.1203 |
|
85 |
+
| 1.152 | 2.24 | 7000 | 1.1199 |
|
86 |
+
| 1.1335 | 2.3 | 7200 | 1.1191 |
|
87 |
+
| 1.1341 | 2.37 | 7400 | 1.1184 |
|
88 |
+
| 1.0963 | 2.43 | 7600 | 1.1180 |
|
89 |
+
| 1.1179 | 2.5 | 7800 | 1.1172 |
|
90 |
+
| 1.1176 | 2.56 | 8000 | 1.1170 |
|
91 |
+
| 1.15 | 2.62 | 8200 | 1.1165 |
|
92 |
+
| 1.073 | 2.69 | 8400 | 1.1164 |
|
93 |
+
| 1.1737 | 2.75 | 8600 | 1.1156 |
|
94 |
+
| 1.1251 | 2.82 | 8800 | 1.1156 |
|
95 |
+
| 1.1548 | 2.88 | 9000 | 1.1154 |
|
96 |
+
| 1.1046 | 2.94 | 9200 | 1.1152 |
|
97 |
+
|
98 |
+
|
99 |
+
### Framework versions
|
100 |
+
|
101 |
+
- Transformers 4.30.0.dev0
|
102 |
+
- Pytorch 2.0.0
|
103 |
+
- Datasets 2.11.0
|
104 |
+
- Tokenizers 0.13.3
|