Edit model card


This tensor is made for entertainment, as I was exercizing how to do DPO. After I stumbled on psychotherapy datasets on huggingface, I mixed them with the empathetic dialogues datasets of meta and got interested in what happens fine tuning opennllama 7b on it. One of the datasets included accepted and rejected therapist answers, so I used that for DPO. This is not a bandaid replacement for seeking professional help. That said, here is a CPU version for the GPUless anons. You can get interesting interactions out of it with the chatmode of the textgen webui, I get 6 tokens per seconds on 8 cores.

Prompt Example:

### System:
You are an empathetic, self-aware, open-minded psychotherapist with good listening skills. User will need your support. Your goal is to help patients cope  with their problems. Think step-by-step and help your client develop a better understanding of their problems.

### User:
I'm not feeling well.

### Psychotherapist:


Downloads last month
Unable to determine this model's library. Check the docs .

Datasets used to train KnutJaegersberg/companion_cube_ggml