Yi-6B-ruozhiba

This model is a fine-tuned version of 01-ai/Yi-6B on the ruozhiba dataset. It achieves the following results on the evaluation set:

• Loss: 4.1672

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
2.0998	1.0	55	2.0215
1.7439	2.0	110	1.8371
1.7017	3.0	165	1.8182
1.3662	4.0	220	1.9196
1.0344	5.0	275	2.2265
0.6959	6.0	330	2.5187
0.4353	7.0	385	2.8303
0.2736	8.0	440	3.1702
0.1598	9.0	495	3.3739
0.117	10.0	550	3.4964
0.0856	11.0	605	3.6468
0.0681	12.0	660	3.8002
0.058	13.0	715	3.9182
0.0532	14.0	770	4.0117
0.0443	15.0	825	4.0919
0.0461	17.0	935	4.1480
0.0465	18.0	990	4.1622
0.049	19.0	1045	4.1667
0.0416	20.0	1100	4.1672

Framework versions

• PEFT 0.7.1
• Transformers 4.36.2
• Pytorch 2.2.2+cu118
• Datasets 2.14.6
• Tokenizers 0.15.2