|
--- |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
base_model: NousResearch/Llama-2-7b-hf |
|
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: 1.6705 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.9724 | 0.0 | 20 | 1.8100 | |
|
| 1.8173 | 0.01 | 40 | 1.7801 | |
|
| 1.8184 | 0.01 | 60 | 1.7671 | |
|
| 1.8725 | 0.01 | 80 | 1.7568 | |
|
| 1.8967 | 0.01 | 100 | 1.7460 | |
|
| 1.8943 | 0.02 | 120 | 1.7172 | |
|
| 1.788 | 0.02 | 140 | 1.7045 | |
|
| 1.8953 | 0.02 | 160 | 1.6986 | |
|
| 1.8262 | 0.02 | 180 | 1.6943 | |
|
| 1.8472 | 0.03 | 200 | 1.6926 | |
|
| 1.8416 | 0.03 | 220 | 1.6896 | |
|
| 1.838 | 0.03 | 240 | 1.6855 | |
|
| 1.7743 | 0.04 | 260 | 1.6806 | |
|
| 1.8562 | 0.04 | 280 | 1.6785 | |
|
| 1.8562 | 0.04 | 300 | 1.6794 | |
|
| 1.8117 | 0.04 | 320 | 1.6783 | |
|
| 1.8193 | 0.05 | 340 | 1.6768 | |
|
| 1.8807 | 0.05 | 360 | 1.6745 | |
|
| 1.7641 | 0.05 | 380 | 1.6738 | |
|
| 1.7738 | 0.05 | 400 | 1.6735 | |
|
| 1.7759 | 0.06 | 420 | 1.6733 | |
|
| 1.7089 | 0.06 | 440 | 1.6721 | |
|
| 1.7984 | 0.06 | 460 | 1.6706 | |
|
| 1.7243 | 0.07 | 480 | 1.6720 | |
|
| 1.9205 | 0.07 | 500 | 1.6705 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.1 |
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
