lmi-ft

This model is a fine-tuned version of TheBloke/Mistral-7B-Instruct-v0.2-GPTQ on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1618

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: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.964	0.9714	17	2.5423
2.5133	1.9714	34	2.2165
2.2166	2.9714	51	2.0860
2.103	3.9714	68	2.0531
1.9895	4.9714	85	2.0560
1.8706	5.9714	102	2.0744
1.7926	6.9714	119	2.1024
1.7225	7.9714	136	2.1234
1.6712	8.9714	153	2.1618
1.5428	9.9714	170	2.1618

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.5.1+cu121
• Datasets 3.3.2
• Tokenizers 0.21.0