Summary4500_M2_1000steps_1e7rate_SFT

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0586

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: 1e-07
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
2.0168	0.0112	50	1.9854
1.3848	0.0224	100	1.3059
0.1374	0.0336	150	0.1142
0.0099	0.0448	200	0.0587
0.0076	0.0559	250	0.0581
0.0073	0.0671	300	0.0580
0.0071	0.0783	350	0.0587
0.0071	0.0895	400	0.0586
0.0069	0.1007	450	0.0589
0.0068	0.1119	500	0.0586
0.0068	0.1231	550	0.0586
0.0067	0.1343	600	0.0588
0.0067	0.1454	650	0.0589
0.0066	0.1566	700	0.0590
0.0066	0.1678	750	0.0587
0.0066	0.1790	800	0.0588
0.0066	0.1902	850	0.0588
0.0066	0.2014	900	0.0586
0.0066	0.2126	950	0.0586
0.0066	0.2238	1000	0.0586

Framework versions

• Transformers 4.42.4
• Pytorch 2.0.0+cu117
• Datasets 2.20.0
• Tokenizers 0.19.1