Yi-6B-zhihu7

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

• Loss: 2.5970

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-06
• train_batch_size: 8
• 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
2.6831	1.0	96	2.6346
2.6684	2.0	192	2.6288
2.687	3.0	288	2.6192
2.6597	4.0	384	2.6109
2.6019	5.0	480	2.6054
2.6118	6.0	576	2.6022
2.7286	7.0	672	2.6001
2.6341	8.0	768	2.5987
2.572	9.0	864	2.5979
2.622	10.0	960	2.5974
2.6404	11.0	1056	2.5972
2.6607	12.0	1152	2.5971
2.5324	13.0	1248	2.5971
2.5472	14.0	1344	2.5970
2.539	15.0	1440	2.5970
2.5757	16.0	1536	2.5971
2.6495	17.0	1632	2.5970
2.5647	18.0	1728	2.5970
2.5605	19.0	1824	2.5970
2.6608	20.0	1920	2.5970

Framework versions

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