shawgpt-ft-trial-15

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: 1.3431

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: 0.0008738460466723728
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 4
• num_epochs: 18
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
7.5176	0.5714	1	4.2400
3.4718	1.7143	3	3.6765
2.8267	2.8571	5	2.6759
2.1885	4.0	7	2.0843
3.423	4.5714	8	1.8487
1.307	5.7143	10	1.5116
1.1129	6.8571	12	1.3905
1.0899	8.0	14	1.3626
2.0976	8.5714	15	1.3542
0.9824	9.7143	17	1.3447
0.4944	10.2857	18	1.3431

Framework versions

• PEFT 0.13.2
• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1