update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- generator
|
| 7 |
+
model-index:
|
| 8 |
+
- name: gpt2-cl-length-sampling-2
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# gpt2-cl-length-sampling-2
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 5.0138
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 0.0005
|
| 39 |
+
- train_batch_size: 64
|
| 40 |
+
- eval_batch_size: 64
|
| 41 |
+
- seed: 42
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: cosine
|
| 44 |
+
- lr_scheduler_warmup_steps: 1000
|
| 45 |
+
- num_epochs: 1
|
| 46 |
+
- mixed_precision_training: Native AMP
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 6.5248 | 0.06 | 500 | 5.9478 |
|
| 53 |
+
| 5.2522 | 0.11 | 1000 | 5.5602 |
|
| 54 |
+
| 4.957 | 0.17 | 1500 | 5.3690 |
|
| 55 |
+
| 4.754 | 0.22 | 2000 | 5.2557 |
|
| 56 |
+
| 4.6118 | 0.28 | 2500 | 5.1715 |
|
| 57 |
+
| 4.4957 | 0.33 | 3000 | 5.1314 |
|
| 58 |
+
| 4.3901 | 0.39 | 3500 | 5.0888 |
|
| 59 |
+
| 4.2957 | 0.44 | 4000 | 5.0603 |
|
| 60 |
+
| 4.2028 | 0.5 | 4500 | 5.0352 |
|
| 61 |
+
| 4.1103 | 0.55 | 5000 | 5.0110 |
|
| 62 |
+
| 4.0272 | 0.61 | 5500 | 4.9948 |
|
| 63 |
+
| 3.9358 | 0.66 | 6000 | 4.9880 |
|
| 64 |
+
| 3.8622 | 0.72 | 6500 | 4.9728 |
|
| 65 |
+
| 3.7912 | 0.77 | 7000 | 4.9692 |
|
| 66 |
+
| 3.739 | 0.83 | 7500 | 4.9586 |
|
| 67 |
+
| 3.6971 | 0.88 | 8000 | 4.9557 |
|
| 68 |
+
| 3.6791 | 0.94 | 8500 | 4.9535 |
|
| 69 |
+
| 3.6652 | 0.99 | 9000 | 4.9530 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- Transformers 4.26.1
|
| 75 |
+
- Pytorch 1.11.0+cu113
|
| 76 |
+
- Datasets 2.13.0
|
| 77 |
+
- Tokenizers 0.13.3
|