|
--- |
|
base_model: NousResearch/Llama-2-7b-hf |
|
datasets: |
|
- generator |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
model-index: |
|
- name: llama2_instruct_generation |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# llama2_instruct_generation |
|
|
|
This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6759 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- training_steps: 500 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 1.9468 | 0.0027 | 20 | 1.8141 | |
|
| 1.8734 | 0.0054 | 40 | 1.7849 | |
|
| 1.8769 | 0.0081 | 60 | 1.7718 | |
|
| 1.8633 | 0.0108 | 80 | 1.7599 | |
|
| 1.8583 | 0.0135 | 100 | 1.7472 | |
|
| 1.8264 | 0.0163 | 120 | 1.7176 | |
|
| 1.8714 | 0.0190 | 140 | 1.7053 | |
|
| 1.831 | 0.0217 | 160 | 1.7012 | |
|
| 1.7957 | 0.0244 | 180 | 1.6947 | |
|
| 1.8613 | 0.0271 | 200 | 1.6934 | |
|
| 1.81 | 0.0298 | 220 | 1.6915 | |
|
| 1.7995 | 0.0325 | 240 | 1.6893 | |
|
| 1.9067 | 0.0352 | 260 | 1.6872 | |
|
| 1.8261 | 0.0379 | 280 | 1.6860 | |
|
| 1.8609 | 0.0406 | 300 | 1.6843 | |
|
| 1.7725 | 0.0433 | 320 | 1.6835 | |
|
| 1.8061 | 0.0461 | 340 | 1.6819 | |
|
| 1.8842 | 0.0488 | 360 | 1.6804 | |
|
| 1.7648 | 0.0515 | 380 | 1.6799 | |
|
| 1.8121 | 0.0542 | 400 | 1.6796 | |
|
| 1.8056 | 0.0569 | 420 | 1.6777 | |
|
| 1.7423 | 0.0596 | 440 | 1.6780 | |
|
| 1.8971 | 0.0623 | 460 | 1.6782 | |
|
| 1.8234 | 0.0650 | 480 | 1.6771 | |
|
| 1.8978 | 0.0677 | 500 | 1.6759 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.13.0 |
|
- Transformers 4.45.0 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.20.0 |
