Chat-Error/Pyg6B-V8P2

1---2license: creativeml-openrail-m3language:4- en5thumbnail:6tags:7- text generation8- conversational9inference: false10---11 12# Pygmalion 6B13 14## Model description15 16Pymalion 6B is a proof-of-concept dialogue model based on EleutherAI's [GPT-J-6B](https://huggingface.co/EleutherAI/gpt-j-6B).17 18**Warning:** This model is **NOT** suitable for use by minors. It **will** output X-rated content under certain circumstances.19 20## Training data21 22The fine-tuning dataset consisted of 56MB of dialogue data gathered from multiple sources, which includes both real _and_ partially machine-generated conversations.23 24## Training procedure25 26Model weights were initialized from the `uft-6b` ConvoGPT model made available in [this commit](https://huggingface.co/hakurei/convogpt/tree/41b67bfddb6cd97070ffddf708e9720c9cb8d224/6b-uft).27 28The model was then further fine-tuned on ~48.5 million tokens for ~5k steps on 4 NVIDIA A40s using DeepSpeed.29 30## Intended use31 32### The easy way33 34We provide a notebook with a Gradio UI for playing around with the model without having to manually format inputs. This notebook can be found [here](https://github.com/PygmalionAI/gradio-ui/blob/master/notebooks/GPU.ipynb).35 36### The manual way37 38The model can be used as a regular text generation model, but it'll perform best if the input prompt adheres to the following format:39 40```41[CHARACTER]'s Persona: [A few sentences about the character you want the model to play]42<START>43[DIALOGUE HISTORY]44You: [Your input message here]45[CHARACTER]:46```47 48Where `[CHARACTER]` is, as you can probably guess, the name of the character you want the model to portray, `<START>` should be used verbatim as a delimiter token to separate persona and scenario data from the dialogue, and `[DIALOGUE HISTORY]` is chat history so the model can have some conversational context to draw from. Ideally it'll be pairs of messages like:49 50```51[CHARACTER]: [some dialogue here]52You: [your response to the dialogue above]53```54 55Apart from chat history, you can also just add example conversations in `[DIALOGUE HISTORY]` to show how the character should speak - ideally at the beginning, so it doesn't get confused as to what's conversation history vs. character definition.56 57## Known issues58 59We haven't played around with the model enough to enumerate them. Feel free to give us some feedback!60