shawgpt-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: 0.4523

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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
2.0963	0.9677	15	1.3918
1.126	2.0	31	0.8829
0.8856	2.9677	46	0.6796
0.6852	4.0	62	0.5936
0.6348	4.9677	77	0.5174
0.5272	6.0	93	0.4824
0.5065	6.9677	108	0.4689
0.4308	8.0	124	0.4598
0.4287	8.9677	139	0.4512
0.3658	9.6774	150	0.4523

Framework versions

• PEFT 0.12.0
• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1