Simon Tang
commited on
Commit
•
816a89d
1
Parent(s):
bbed8e6
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- anli
|
| 7 |
+
model-index:
|
| 8 |
+
- name: sft-trl
|
| 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 |
+
# sft-trl
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [EleutherAI/gpt-j-6b](https://huggingface.co/EleutherAI/gpt-j-6b) on the anli dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.6837
|
| 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: 5e-05
|
| 39 |
+
- train_batch_size: 8
|
| 40 |
+
- eval_batch_size: 8
|
| 41 |
+
- seed: 42
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 100
|
| 45 |
+
- num_epochs: 3.0
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 51 |
+
| 2.1859 | 0.2 | 1000 | 2.0019 |
|
| 52 |
+
| 1.1342 | 0.4 | 2000 | 1.5352 |
|
| 53 |
+
| 0.8871 | 0.6 | 3000 | 1.0923 |
|
| 54 |
+
| 0.591 | 0.8 | 4000 | 0.8373 |
|
| 55 |
+
| 0.3165 | 1.0 | 5000 | 0.7852 |
|
| 56 |
+
| 0.3205 | 1.2 | 6000 | 0.7531 |
|
| 57 |
+
| 0.2338 | 1.4 | 7000 | 0.7155 |
|
| 58 |
+
| 0.2922 | 1.6 | 8000 | 0.6837 |
|
| 59 |
+
| 0.2427 | 1.8 | 9000 | 0.6837 |
|
| 60 |
+
| 0.2036 | 2.0 | 10000 | 0.6837 |
|
| 61 |
+
| 0.1205 | 2.2 | 11000 | 0.6837 |
|
| 62 |
+
| 0.2275 | 2.4 | 12000 | 0.6837 |
|
| 63 |
+
| 0.1094 | 2.6 | 13000 | 0.6837 |
|
| 64 |
+
| 0.1507 | 2.8 | 14000 | 0.6837 |
|
| 65 |
+
| 0.1739 | 3.0 | 15000 | 0.6837 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.30.2
|
| 71 |
+
- Pytorch 2.0.1+cu117
|
| 72 |
+
- Datasets 2.13.1
|
| 73 |
+
- Tokenizers 0.13.3
|