metadata
license: llama3
base_model: meta-llama/Meta-Llama-3-8B-Instruct
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Summary4500_L3_600steps_1e7rate_SFT
results: []
Summary4500_L3_600steps_1e7rate_SFT
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7224
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: 1e-07
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 600
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.2833 | 0.0112 | 50 | 2.2995 |
| 2.2064 | 0.0224 | 100 | 2.2092 |
| 2.0654 | 0.0336 | 150 | 2.0720 |
| 1.9437 | 0.0448 | 200 | 1.9576 |
| 1.8484 | 0.0559 | 250 | 1.8656 |
| 1.7789 | 0.0671 | 300 | 1.7963 |
| 1.7333 | 0.0783 | 350 | 1.7572 |
| 1.7091 | 0.0895 | 400 | 1.7327 |
| 1.6971 | 0.1007 | 450 | 1.7244 |
| 1.6949 | 0.1119 | 500 | 1.7222 |
| 1.6943 | 0.1231 | 550 | 1.7224 |
| 1.694 | 0.1343 | 600 | 1.7224 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1