|
--- |
|
license: apache-2.0 |
|
base_model: distilbert/distilbert-base-uncased |
|
tags: |
|
- trl |
|
- reward-trainer |
|
- generated_from_trainer |
|
datasets: |
|
- hdfs_rlhf_log_summary_dataset |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: log_sage_reward_model |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: hdfs_rlhf_log_summary_dataset |
|
type: hdfs_rlhf_log_summary_dataset |
|
config: default |
|
split: None |
|
args: default |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 1.0 |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# log_sage_reward_model |
|
|
|
This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on the hdfs_rlhf_log_summary_dataset dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1669 |
|
- Accuracy: 1.0 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 1.41e-05 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 24 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 96 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 40 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 1.0 | 1 | 0.6950 | 0.5 | |
|
| No log | 2.0 | 2 | 0.6896 | 1.0 | |
|
| No log | 3.0 | 3 | 0.6843 | 1.0 | |
|
| No log | 4.0 | 4 | 0.6789 | 1.0 | |
|
| No log | 5.0 | 5 | 0.6735 | 1.0 | |
|
| No log | 6.0 | 6 | 0.6671 | 1.0 | |
|
| No log | 7.0 | 7 | 0.6597 | 1.0 | |
|
| No log | 8.0 | 8 | 0.6510 | 1.0 | |
|
| No log | 9.0 | 9 | 0.6403 | 1.0 | |
|
| 0.0839 | 10.0 | 10 | 0.6275 | 1.0 | |
|
| 0.0839 | 11.0 | 11 | 0.6130 | 1.0 | |
|
| 0.0839 | 12.0 | 12 | 0.5955 | 1.0 | |
|
| 0.0839 | 13.0 | 13 | 0.5747 | 1.0 | |
|
| 0.0839 | 14.0 | 14 | 0.5508 | 1.0 | |
|
| 0.0839 | 15.0 | 15 | 0.5250 | 1.0 | |
|
| 0.0839 | 16.0 | 16 | 0.4984 | 1.0 | |
|
| 0.0839 | 17.0 | 17 | 0.4698 | 1.0 | |
|
| 0.0839 | 18.0 | 18 | 0.4413 | 1.0 | |
|
| 0.0839 | 19.0 | 19 | 0.4121 | 1.0 | |
|
| 0.0658 | 20.0 | 20 | 0.3850 | 1.0 | |
|
| 0.0658 | 21.0 | 21 | 0.3604 | 1.0 | |
|
| 0.0658 | 22.0 | 22 | 0.3384 | 1.0 | |
|
| 0.0658 | 23.0 | 23 | 0.3186 | 1.0 | |
|
| 0.0658 | 24.0 | 24 | 0.2995 | 1.0 | |
|
| 0.0658 | 25.0 | 25 | 0.2823 | 1.0 | |
|
| 0.0658 | 26.0 | 26 | 0.2664 | 1.0 | |
|
| 0.0658 | 27.0 | 27 | 0.2516 | 1.0 | |
|
| 0.0658 | 28.0 | 28 | 0.2384 | 1.0 | |
|
| 0.0658 | 29.0 | 29 | 0.2260 | 1.0 | |
|
| 0.0346 | 30.0 | 30 | 0.2149 | 1.0 | |
|
| 0.0346 | 31.0 | 31 | 0.2054 | 1.0 | |
|
| 0.0346 | 32.0 | 32 | 0.1971 | 1.0 | |
|
| 0.0346 | 33.0 | 33 | 0.1898 | 1.0 | |
|
| 0.0346 | 34.0 | 34 | 0.1838 | 1.0 | |
|
| 0.0346 | 35.0 | 35 | 0.1787 | 1.0 | |
|
| 0.0346 | 36.0 | 36 | 0.1746 | 1.0 | |
|
| 0.0346 | 37.0 | 37 | 0.1714 | 1.0 | |
|
| 0.0346 | 38.0 | 38 | 0.1691 | 1.0 | |
|
| 0.0346 | 39.0 | 39 | 0.1676 | 1.0 | |
|
| 0.021 | 40.0 | 40 | 0.1669 | 1.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
