--- license: apache-2.0 library_name: peft tags: - trl - sft - generated_from_trainer base_model: mistralai/Mistral-7B-Instruct-v0.1 model-index: - name: mistral-7b-instruct-connections results: [] --- [Visualize in Weights & Biases](https://wandb.ai/emilycs229/huggingface/runs/405rlxb2) [Visualize in Weights & Biases](https://wandb.ai/emilycs229/huggingface/runs/405rlxb2) [Visualize in Weights & Biases](https://wandb.ai/emilycs229/huggingface/runs/405rlxb2) [Visualize in Weights & Biases](https://wandb.ai/emilycs229/huggingface/runs/405rlxb2) [Visualize in Weights & Biases](https://wandb.ai/emilycs229/224n_connections/runs/vz9otsu5) [Visualize in Weights & Biases](https://wandb.ai/emilycs229/224n_connections/runs/k7e9e0v3) # mistral-7b-instruct-connections This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.6526 ## 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: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: constant - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | No log | 0.9855 | 17 | 0.5535 | | 0.8187 | 1.9710 | 34 | 0.5099 | | 0.4261 | 2.9565 | 51 | 0.5141 | | 0.4261 | 4.0 | 69 | 0.5708 | | 0.2769 | 4.9275 | 85 | 0.6526 | ### Framework versions - PEFT 0.11.1 - Transformers 4.42.0.dev0 - Pytorch 2.3.0+cu121 - Datasets 2.19.1 - Tokenizers 0.19.1