library_name: transformers
license: other
base_model: meta-llama/Meta-Llama-3-8B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: no_explain
results: []
no_explain
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the chess_explain_noexplain_00, the chess_explain_noexplain_01, the chess_explain_noexplain_02, the chess_explain_noexplain_03, the chess_explain_noexplain_04, the chess_explain_noexplain_05, the chess_explain_noexplain_06, the chess_explain_noexplain_07, the chess_explain_noexplain_08, the chess_explain_noexplain_09, the chess_explain_noexplain_10, the chess_explain_noexplain_11, the chess_explain_noexplain_12, the chess_explain_noexplain_13 and the chess_explain_noexplain_14 datasets. It achieves the following results on the evaluation set:
- Loss: 0.0932
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: 5e-06
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 1024
- total_eval_batch_size: 512
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0429 | 0.8010 | 1000 | 0.0422 |
| 0.0329 | 1.6015 | 2000 | 0.0336 |
| 0.0275 | 2.4021 | 3000 | 0.0297 |
| 0.0202 | 3.2026 | 4000 | 0.0292 |
| 0.0194 | 4.0032 | 5000 | 0.0294 |
| 0.0119 | 4.8042 | 6000 | 0.0311 |
| 0.0048 | 5.6047 | 7000 | 0.0439 |
| 0.0013 | 6.4053 | 8000 | 0.0538 |
| 0.0004 | 7.2058 | 9000 | 0.0670 |
| 0.0003 | 8.0064 | 10000 | 0.0698 |
| 0.0 | 8.8074 | 11000 | 0.0894 |
| 0.0 | 9.6079 | 12000 | 0.0931 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.6.0+cu124
- Datasets 2.21.0
- Tokenizers 0.21.0