metadata
base_model: meta-llama/Meta-Llama-3.1-70B
library_name: peft
license: llama3.1
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: english_tm_model
results: []
english_tm_model
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-70B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2883
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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 25
- training_steps: 300
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.75 | 0.32 | 50 | 1.3528 |
| 1.338 | 0.64 | 100 | 1.3195 |
| 1.3357 | 0.96 | 150 | 1.2975 |
| 1.2553 | 1.28 | 200 | 1.2956 |
| 1.2688 | 1.6 | 250 | 1.2909 |
| 1.2363 | 1.92 | 300 | 1.2883 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.0+cu118
- Datasets 2.21.0
- Tokenizers 0.19.1