CAS-SIAT-XinHai/CPsyCounX

CPsyCounX

This model is a fine-tuned version of internlm/internlm2-chat-7b on the CPsyCounD dataset.

Model description

CPsyCounX is a large language model designed for Chinese Psychological Counseling. Paper: CPsyCoun.

Training and evaluation data

• —Train: CPsyCounD
• —Evaluate: CPsyCounE

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 1e-06
• —trainbatchsize: 4
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 4
• —gradientaccumulationsteps: 28
• —totaltrainbatch_size: 448
• —totalevalbatch_size: 32
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: cosine
• —num_epochs: 9.0
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.37.1
• —Pytorch 2.1.2+cu121
• —Datasets 2.16.1
• —Tokenizers 0.15.1