metadata
base_model: meta-llama/Meta-Llama-3.1-8B
datasets:
- llama-duo/synth_classification_dataset_dedup
library_name: peft
license: llama3.1
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: llama3.1-8b-classification-gpt4o-100k
results: []
llama3.1-8b-classification-gpt4o-100k
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the llama-duo/synth_classification_dataset_dedup dataset. It achieves the following results on the evaluation set:
- Loss: 3.0330
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: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2062 | 1.0 | 296 | 1.6781 |
| 1.1339 | 2.0 | 592 | 1.6897 |
| 1.0779 | 3.0 | 888 | 1.7536 |
| 1.0043 | 4.0 | 1184 | 1.8225 |
| 0.9288 | 5.0 | 1480 | 2.0044 |
| 0.8437 | 6.0 | 1776 | 2.1710 |
| 0.7654 | 7.0 | 2072 | 2.4080 |
| 0.7117 | 8.0 | 2368 | 2.6554 |
| 0.6916 | 9.0 | 2664 | 2.9172 |
| 0.6652 | 10.0 | 2960 | 3.0330 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1