|
|
--- |
|
|
library_name: peft |
|
|
license: llama3.1 |
|
|
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: Llama-Instruct-8B |
|
|
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. --> |
|
|
|
|
|
# Llama-Instruct-8B |
|
|
|
|
|
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.2965 |
|
|
|
|
|
## 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: 4 |
|
|
- eval_batch_size: 4 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 100 |
|
|
- num_epochs: 4 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 2.1139 | 0.1144 | 50 | 1.8861 | |
|
|
| 1.3487 | 0.2288 | 100 | 0.6872 | |
|
|
| 0.4797 | 0.3432 | 150 | 0.4065 | |
|
|
| 0.3914 | 0.4577 | 200 | 0.3877 | |
|
|
| 0.3808 | 0.5721 | 250 | 0.3773 | |
|
|
| 0.3682 | 0.6865 | 300 | 0.3622 | |
|
|
| 0.3539 | 0.8009 | 350 | 0.3459 | |
|
|
| 0.3333 | 0.9153 | 400 | 0.3344 | |
|
|
| 0.3278 | 1.0297 | 450 | 0.3261 | |
|
|
| 0.3227 | 1.1442 | 500 | 0.3215 | |
|
|
| 0.3182 | 1.2586 | 550 | 0.3185 | |
|
|
| 0.315 | 1.3730 | 600 | 0.3156 | |
|
|
| 0.3117 | 1.4874 | 650 | 0.3142 | |
|
|
| 0.3108 | 1.6018 | 700 | 0.3122 | |
|
|
| 0.3083 | 1.7162 | 750 | 0.3113 | |
|
|
| 0.3086 | 1.8307 | 800 | 0.3089 | |
|
|
| 0.3083 | 1.9451 | 850 | 0.3075 | |
|
|
| 0.3054 | 2.0595 | 900 | 0.3070 | |
|
|
| 0.3043 | 2.1739 | 950 | 0.3054 | |
|
|
| 0.301 | 2.2883 | 1000 | 0.3040 | |
|
|
| 0.3023 | 2.4027 | 1050 | 0.3034 | |
|
|
| 0.2988 | 2.5172 | 1100 | 0.3025 | |
|
|
| 0.2988 | 2.6316 | 1150 | 0.3023 | |
|
|
| 0.2988 | 2.7460 | 1200 | 0.3007 | |
|
|
| 0.2987 | 2.8604 | 1250 | 0.3002 | |
|
|
| 0.2974 | 2.9748 | 1300 | 0.2999 | |
|
|
| 0.2966 | 3.0892 | 1350 | 0.2991 | |
|
|
| 0.2966 | 3.2037 | 1400 | 0.2988 | |
|
|
| 0.2963 | 3.3181 | 1450 | 0.2981 | |
|
|
| 0.295 | 3.4325 | 1500 | 0.2979 | |
|
|
| 0.2931 | 3.5469 | 1550 | 0.2974 | |
|
|
| 0.2944 | 3.6613 | 1600 | 0.2972 | |
|
|
| 0.2937 | 3.7757 | 1650 | 0.2967 | |
|
|
| 0.2904 | 3.8902 | 1700 | 0.2965 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.14.0 |
|
|
- Transformers 4.50.3 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 3.5.0 |
|
|
- Tokenizers 0.21.1 |
