metadata
license: other
base_model: meta-llama/Meta-Llama-3-8B
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: C018_random_sample_llama3-8b-base_instruct_20240504_182259
results: []
C018_random_sample_llama3-8b-base_instruct_20240504_182259
This model is a fine-tuned version of ./output/training_results/C018_random_sample_llama3-8b-base_pretrain_20240504_182259/ on the instructions_curated dataset. It achieves the following results on the evaluation set:
- Loss: 0.8423
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: 1.5e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 64
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_steps: 20
- num_epochs: 4.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8108 | 0.4167 | 20 | 0.8423 |
| 0.7995 | 0.8333 | 40 | 0.8555 |
| 0.4526 | 1.25 | 60 | 0.8816 |
| 0.4663 | 1.6667 | 80 | 0.8521 |
| 0.3927 | 2.0833 | 100 | 0.8507 |
| 0.4017 | 2.5 | 120 | 0.8561 |
| 0.368 | 2.9167 | 140 | 0.8608 |
| 0.3677 | 3.3333 | 160 | 0.8647 |
| 0.3635 | 3.75 | 180 | 0.8676 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0
- Datasets 2.19.0
- Tokenizers 0.19.1