ghc-mistralai-Mistral-7B-v0.1-intra-dataset-frequency-model-pairwise-mse-cycle1

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6466

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss
2.4036	0.05	32	2.3614
2.2287	1.05	64	2.2284
2.0581	2.05	96	2.0611
1.8523	3.05	128	1.8729
1.5713	4.05	160	1.6971
1.3776	5.05	192	1.5329
1.2073	6.05	224	1.3806
1.0491	7.05	256	1.2418
0.9345	8.05	288	1.1253
0.8332	9.05	320	1.0236
0.7533	10.05	352	0.9339
0.6942	11.05	384	0.8607
0.6392	12.05	416	0.8039
0.5871	13.05	448	0.7580
0.5632	14.05	480	0.7216
0.5303	15.05	512	0.6928
0.5132	16.05	544	0.6740
0.494	17.05	576	0.6599
0.4862	18.05	608	0.6517
0.489	19.05	640	0.6476
0.484	20.05	672	0.6466
0.4894	21.05	704	0.6466
0.4809	22.05	736	0.6466
0.4751	23.05	768	0.6466
0.4779	24.05	800	0.6466
0.4731	25.05	832	0.6466
0.4784	26.05	864	0.6466
0.4768	27.05	896	0.6466

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.1+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0