|
--- |
|
license: llama2 |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
base_model: meta-llama/Llama-2-7b-hf |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
model-index: |
|
- name: llama2-7B_MT |
|
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. --> |
|
|
|
# llama2-7B_MT |
|
|
|
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7960 |
|
- Accuracy: 0.8317 |
|
- Precision: 0.8541 |
|
- Recall: 0.8 |
|
- F1 score: 0.8262 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 score | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:| |
|
| 0.693 | 0.25 | 200 | 0.6029 | 0.7617 | 0.8398 | 0.6467 | 0.7307 | |
|
| 0.5602 | 0.5 | 400 | 0.6130 | 0.7733 | 0.8661 | 0.6467 | 0.7405 | |
|
| 0.5364 | 0.75 | 600 | 0.4880 | 0.785 | 0.7714 | 0.81 | 0.7902 | |
|
| 0.5136 | 1.0 | 800 | 0.5408 | 0.7717 | 0.8327 | 0.68 | 0.7486 | |
|
| 0.3945 | 1.25 | 1000 | 0.6649 | 0.7683 | 0.9215 | 0.5867 | 0.7169 | |
|
| 0.3574 | 1.5 | 1200 | 0.5797 | 0.7767 | 0.7413 | 0.85 | 0.7919 | |
|
| 0.3927 | 1.75 | 1400 | 0.4764 | 0.8267 | 0.8333 | 0.8167 | 0.8249 | |
|
| 0.3584 | 2.0 | 1600 | 0.4186 | 0.8267 | 0.8475 | 0.7967 | 0.8213 | |
|
| 0.2488 | 2.25 | 1800 | 0.4973 | 0.8317 | 0.8755 | 0.7733 | 0.8212 | |
|
| 0.2519 | 2.5 | 2000 | 0.5590 | 0.8217 | 0.8814 | 0.7433 | 0.8065 | |
|
| 0.2424 | 2.75 | 2200 | 0.6088 | 0.8217 | 0.8587 | 0.77 | 0.8120 | |
|
| 0.2517 | 3.0 | 2400 | 0.5793 | 0.8317 | 0.8964 | 0.75 | 0.8167 | |
|
| 0.1178 | 3.25 | 2600 | 0.6630 | 0.8183 | 0.8498 | 0.7733 | 0.8098 | |
|
| 0.1058 | 3.5 | 2800 | 0.9330 | 0.8167 | 0.8958 | 0.7167 | 0.7963 | |
|
| 0.098 | 3.75 | 3000 | 0.7077 | 0.82 | 0.8137 | 0.83 | 0.8218 | |
|
| 0.0875 | 4.0 | 3200 | 0.6751 | 0.82 | 0.8288 | 0.8067 | 0.8176 | |
|
| 0.0251 | 4.25 | 3400 | 0.7202 | 0.8283 | 0.8339 | 0.82 | 0.8269 | |
|
| 0.0202 | 4.5 | 3600 | 0.7859 | 0.83 | 0.8587 | 0.79 | 0.8229 | |
|
| 0.0196 | 4.75 | 3800 | 0.8298 | 0.8333 | 0.8650 | 0.79 | 0.8258 | |
|
| 0.0174 | 5.0 | 4000 | 0.7960 | 0.8317 | 0.8541 | 0.8 | 0.8262 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.11.1 |
|
- Transformers 4.44.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
