Zeroshot-3.2.3-Mistral-7B-pipeline-config
This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.7373
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 342
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.0342 | 0.81 | 50 | 0.8600 |
0.8239 | 1.62 | 100 | 0.8089 |
0.7904 | 2.43 | 150 | 0.7841 |
0.7596 | 3.24 | 200 | 0.7662 |
0.7394 | 4.05 | 250 | 0.7511 |
0.7133 | 4.86 | 300 | 0.7373 |
Framework versions
- PEFT 0.7.1
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.16.1
- Tokenizers 0.15.1
- Downloads last month
- 0
Model tree for Weni/Zeroshot-3.2.3-Mistral-7B-pipeline-config
Base model
mistralai/Mistral-7B-v0.1
Finetuned
HuggingFaceH4/zephyr-7b-beta