shawgpt-ft-epoch-13

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: 2.9662

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 13
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
8.5018	0.5714	1	4.2401
8.5334	1.5714	2	4.1568
8.0825	2.5714	3	3.9663
7.7613	3.5714	4	3.7876
7.3597	4.5714	5	3.6281
7.0923	5.5714	6	3.4838
6.8333	6.5714	7	3.3548
6.5917	7.5714	8	3.2436
6.4073	8.5714	9	3.1514
6.2985	9.5714	10	3.0779
6.0792	10.5714	11	3.0223
6.0114	11.5714	12	2.9852
3.0019	12.5714	13	2.9662

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0