---
license: mit
base_model: TheBloke/zephyr-7B-alpha-GPTQ
tags:
- trl
- sft
- generated_from_trainer
- peft
- gptq
model-index:
- name: thesa
  results: []
language:
- en
datasets:
- loaiabdalslam/counselchat
pipeline_tag: text-generation
# widget:
# - text: "<|system|>You are a therapist helping patients.<|user|>I'm fighting with my boyfriend and he's not talking to me. I don't know what to do<|assistant|>"
#   example_title: "Example 3"
---

# Thesa

Thesa is an experimental project of a therapy chatbot trained on mental health data and fine-tuned with the Zephyr GPTQ model that uses quantization to decrease high computatinal and storage costs.

## Model description

- Fine-tuned from [TheBloke/zephyr-7B-alpha-GPTQ](https://huggingface.co/TheBloke/zephyr-7B-alpha-GPTQ)

## Intended uses & limitations

The intended use is experimental.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- training_steps: 250
- mixed_precision_training: Native AMP


### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1

## More info
More info at https://github.com/johnhandleyd/thesa