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: 2.3113

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
4.699	1.0	2	4.4836
4.4466	2.0	4	3.9976
3.9547	3.0	6	3.5964
3.5685	4.0	8	3.2800
3.2599	5.0	10	3.0156
2.9994	6.0	12	2.7879
2.7688	7.0	14	2.5980
2.5856	8.0	16	2.4522
2.4501	9.0	18	2.3550
2.3649	10.0	20	2.3113

Framework versions

• PEFT 0.11.1
• Transformers 4.41.2
• Pytorch 2.1.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1