Yi-6B-ruozhiba2

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: 4.3202

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
2.7303	1.0	55	2.2761
2.0833	2.0	110	1.9984
1.8367	3.0	165	1.9820
1.3057	5.0	275	2.2829
0.972	6.0	330	2.6339
0.6803	7.0	385	2.9009
0.4926	8.0	440	3.1734
0.3161	9.0	495	3.3547
0.2503	10.0	550	3.6204
0.1914	11.0	605	3.7289
0.1431	12.0	660	3.8908
0.1286	13.0	715	4.0403
0.1169	14.0	770	4.1426
0.1122	15.0	825	4.1888
0.1046	16.0	880	4.2556
0.1045	17.0	935	4.2806
0.1203	18.0	990	4.3127
0.1097	19.0	1045	4.3194
0.1087	20.0	1100	4.3202

Framework versions

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