Downloads · 30 days
11
2% of all-time downloads
i5-8300h/gemma3-4b-psychology-dpo
gemma3-4b-psychology-dpo is a text generation model from i5-8300h. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as gemma.
DPO and LoRA PeFT finetuned version of Gemma 3 4B for response tone adjustment
Downloads · 30 days
11
2% of all-time downloads
All-time downloads
465
Public
Parameters
4.3B
8.6 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8.6 GB · 100%
From the Hugging Face model README
DPO and LoRA PeFT finetuned version of Gemma 3 4B for response tone adjustment
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
Psychologically aligned chatbot tasks
Psychologically aligned chatbot tasks
Do not use in any safety critical or medically critical applications.
Use the code below to get started with the model.
This repo contains merged weights.
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "i5-8300h/gemma3-4b-psychology-dpo"
tok = AutoTokenizer.from_pretrained(model_name, use_fast=True)
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
https://huggingface.co/datasets/entfane/psychotherapy-dpo
Custom collator pads prompt/chosen/rejected groups independently using tokenizer.pad and ensures ID tensors are torch.long.
Python 3.11.8
Environment
Ram Sundar Radhakrishnan
Hugging Face Profile