metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: cognitivecomputations/dolphin-2.2.1-mistral-7b
datasets:
- generator
model-index:
- name: mistral_instruct_generation
results: []
mistral_instruct_generation
This model is a fine-tuned version of cognitivecomputations/dolphin-2.2.1-mistral-7b on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 1.5008
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: 3
- 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
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9388 | 0.1869 | 20 | 1.8326 |
| 1.7574 | 0.3738 | 40 | 1.6963 |
| 1.6736 | 0.5607 | 60 | 1.6514 |
| 1.6703 | 0.7477 | 80 | 1.6321 |
| 1.6708 | 0.9346 | 100 | 1.6113 |
| 1.6049 | 1.1215 | 120 | 1.5912 |
| 1.6446 | 1.3084 | 140 | 1.5891 |
| 1.5222 | 1.4953 | 160 | 1.5691 |
| 1.6022 | 1.6822 | 180 | 1.5599 |
| 1.4326 | 1.8692 | 200 | 1.5512 |
| 1.4876 | 2.0561 | 220 | 1.5548 |
| 1.4821 | 2.2430 | 240 | 1.5379 |
| 1.3868 | 2.4299 | 260 | 1.5326 |
| 1.5277 | 2.6168 | 280 | 1.5286 |
| 1.5205 | 2.8037 | 300 | 1.5084 |
| 1.4125 | 2.9907 | 320 | 1.5008 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1