|
--- |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
base_model: NousResearch/Llama-2-7b-hf |
|
datasets: |
|
- generator |
|
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: 1.6728 |
|
|
|
## 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 |
|
- lr_scheduler_warmup_steps: 0.03 |
|
- training_steps: 500 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 1.8965 | 0.0027 | 20 | 1.8107 | |
|
| 1.8264 | 0.0054 | 40 | 1.7808 | |
|
| 1.9263 | 0.0081 | 60 | 1.7669 | |
|
| 1.8825 | 0.0108 | 80 | 1.7538 | |
|
| 1.8391 | 0.0135 | 100 | 1.7324 | |
|
| 1.8253 | 0.0163 | 120 | 1.7068 | |
|
| 1.9297 | 0.0190 | 140 | 1.7029 | |
|
| 1.8338 | 0.0217 | 160 | 1.6985 | |
|
| 1.7641 | 0.0244 | 180 | 1.6939 | |
|
| 1.7714 | 0.0271 | 200 | 1.6916 | |
|
| 1.715 | 0.0298 | 220 | 1.6886 | |
|
| 1.7786 | 0.0325 | 240 | 1.6866 | |
|
| 1.716 | 0.0352 | 260 | 1.6860 | |
|
| 1.899 | 0.0379 | 280 | 1.6857 | |
|
| 1.8455 | 0.0406 | 300 | 1.6851 | |
|
| 1.6802 | 0.0433 | 320 | 1.6836 | |
|
| 1.7304 | 0.0461 | 340 | 1.6818 | |
|
| 1.8154 | 0.0488 | 360 | 1.6793 | |
|
| 1.8813 | 0.0515 | 380 | 1.6793 | |
|
| 1.8423 | 0.0542 | 400 | 1.6789 | |
|
| 1.7785 | 0.0569 | 420 | 1.6767 | |
|
| 1.8999 | 0.0596 | 440 | 1.6763 | |
|
| 1.7594 | 0.0623 | 460 | 1.6752 | |
|
| 1.7182 | 0.0650 | 480 | 1.6745 | |
|
| 1.8259 | 0.0677 | 500 | 1.6728 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.11.1 |
|
- Transformers 4.41.1 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |