mcdc-test-mistral

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5090
• Perplexity: 1.6636

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Perplexity
1.249	0.2632	5	1.1492	3.1557
0.8949	0.5263	10	0.9671	2.6304
0.9403	0.7895	15	0.8533	2.3474
0.8411	1.0526	20	0.7721	2.1643
0.5975	1.3158	25	0.7005	2.0148
0.7067	1.5789	30	0.6430	1.9021
0.564	1.8421	35	0.5939	1.8111
0.5615	2.1053	40	0.5566	1.7448
0.5646	2.3684	45	0.5321	1.7025
0.4684	2.6316	50	0.5167	1.6764
0.5739	2.8947	55	0.5090	1.6636

Framework versions

• PEFT 0.14.0
• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0