metadata
base_model: mistralai/Mistral-7B-v0.3
datasets:
- llama-duo/synth_closed_qa_dataset_dedup
library_name: peft
license: apache-2.0
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: mistral_7b_0_3-closedqa-gpt4o-100k
results: []
mistral_7b_0_3-closedqa-gpt4o-100k
This model is a fine-tuned version of mistralai/Mistral-7B-v0.3 on the llama-duo/synth_closed_qa_dataset_dedup dataset. It achieves the following results on the evaluation set:
- Loss: 1.7580
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: 8
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- total_eval_batch_size: 32
- 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 |
|---|---|---|---|
| 0.7127 | 1.0 | 146 | 1.7010 |
| 0.6581 | 2.0 | 292 | 1.6853 |
| 0.6323 | 3.0 | 438 | 1.6891 |
| 0.6067 | 4.0 | 584 | 1.7035 |
| 0.591 | 5.0 | 730 | 1.7226 |
| 0.5752 | 6.0 | 876 | 1.7259 |
| 0.5603 | 7.0 | 1022 | 1.7422 |
| 0.5632 | 8.0 | 1168 | 1.7483 |
| 0.5497 | 9.0 | 1314 | 1.7570 |
| 0.5353 | 10.0 | 1460 | 1.7580 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.2.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1