metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.2
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Summary4500_M2_400steps_1e8rate_SFT
results: []
Summary4500_M2_400steps_1e8rate_SFT
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.9675
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: 1e-08
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 400
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9563 | 0.0447 | 50 | 1.9699 |
| 1.9559 | 0.0895 | 100 | 1.9696 |
| 1.9642 | 0.1342 | 150 | 1.9686 |
| 1.9621 | 0.1790 | 200 | 1.9673 |
| 1.9548 | 0.2237 | 250 | 1.9676 |
| 1.9541 | 0.2685 | 300 | 1.9678 |
| 1.9743 | 0.3132 | 350 | 1.9675 |
| 1.964 | 0.3579 | 400 | 1.9675 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1