shawgpt-ft-epoch-25

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: 1.9760

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: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
8.5018	0.5714	1	4.2401
8.5321	1.5714	2	4.1560
8.0764	2.5714	3	3.9643
7.7407	3.5714	4	3.7754
7.3053	4.5714	5	3.5978
6.9918	5.5714	6	3.4268
6.6703	6.5714	7	3.2631
6.3584	7.5714	8	3.1110
6.0966	8.5714	9	2.9737
5.8804	9.5714	10	2.8524
5.563	10.5714	11	2.7455
5.3723	11.5714	12	2.6487
5.1872	12.5714	13	2.5584
4.994	13.5714	14	2.4747
4.8122	14.5714	15	2.3941
4.6556	15.5714	16	2.3245
4.5241	16.5714	17	2.2687
4.4362	17.5714	18	2.2153
4.314	18.5714	19	2.1626
4.1867	19.5714	20	2.1120
4.0063	20.5714	21	2.0663
3.9645	21.5714	22	2.0290
3.8819	22.5714	23	2.0020
3.8298	23.5714	24	1.9846
1.9396	24.5714	25	1.9760

Framework versions

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