Mistral_Ch_Text_Grand

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4386

Model description

The model convert a user input into a detailed technical description

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• 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: 50
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9051	0.2886	50	0.8378
0.6498	0.5772	100	0.6354
0.559	0.8658	150	0.5632
0.4766	1.1501	200	0.5298
0.4575	1.4387	250	0.4946
0.4517	1.7273	300	0.4729
0.4278	2.0115	350	0.4606
0.4048	2.3001	400	0.4517
0.4137	2.5887	450	0.4427
0.3779	2.8773	500	0.4386

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1