--- library_name: peft license: apache-2.0 base_model: unsloth/mistral-7b-bnb-4bit tags: - unsloth - generated_from_trainer model-index: - name: english-hindi-colloquial-translator results: [] datasets: - vanshikasundrani/english-hindi-colloquial-dataset - cfilt/iitb-english-hindi --- # english-hindi-colloquial-translator This model is a fine-tuned version of [unsloth/mistral-7b-bnb-4bit](https://huggingface.co/unsloth/mistral-7b-bnb-4bit) on the None dataset. It achieves the following results on the evaluation set: - Loss: 8.7228 ## Model description Model to translate the English to Hindi Language ## Intended uses & limitations More information needed ## Training and evaluation data Base pretrained dataset: Training and Evaluation Data is present in: vanshikasundrani/english-hindi-colloquial-dataset prompt = f"""### Human: You are a Hindi colloquial language translator. Translate the following English text to Hindi colloquial language (spoken Hindi). Here are some examples: "See you later?" -> ""बाद में मिलते हैं?" "How are you?" -> "कैसे हो?" "What's your plan?" -> "तुम्हारा क्या प्लान है?" ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 12 - eval_batch_size: 12 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 2 - num_epochs: 8 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 4.6198 | 5.9701 | 400 | 8.7228 | ### Framework versions - PEFT 0.14.0 - Transformers 4.48.3 - Pytorch 2.6.0+cu124 - Datasets 3.3.2 - Tokenizers 0.21.0