metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
datasets:
- generator
model-index:
- name: news-summarization-finetuned-mistral-7b
results: []
news-summarization-finetuned-mistral-7b
This model is a fine-tuned version of unsloth/llama-3-8b-Instruct-bnb-4bit on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 2.4730
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 0.03
- training_steps: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.2259 | 0.2985 | 20 | 2.4012 |
| 1.9923 | 0.5970 | 40 | 2.2957 |
| 1.9638 | 0.8955 | 60 | 2.2907 |
| 1.5504 | 1.1940 | 80 | 2.3691 |
| 1.4113 | 1.4925 | 100 | 2.4730 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.16.0
- Tokenizers 0.19.1