PPO_Pygway-6b-Mix

KoboldAI/PPO_Pygway-6b-Mix

PPO_Pygway-6b-Mix is a text generation model from KoboldAI. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.

This is a merged model, using a weighted parameter blend strategy at a (20:20:60) ratio between the models:

Downloads · 30 days

133

0% of all-time downloads

All-time downloads

142K

Public

Parameters

6.2B

24.4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin12.2 GB · 50%

Parameter types

How the weights are stored.

F166.1B · 98%

Try a prompt

Spaces

31 linked

Model card

From the Hugging Face model README

GPT-J 6B - PPO_Pygway Mix

Model description

This is a merged model, using a weighted parameter blend strategy at a (20:20:60) ratio between the models:

[20%] - KoboldAI/GPT-J-6B-Janeway: https://huggingface.co/KoboldAI/GPT-J-6B-Janeway
[20%] - reciprocate/ppo_hh_gpt-j: https://huggingface.co/reciprocate/ppo_hh_gpt-j
[60%] - Pygmalion/Pygmalion-6b: https://huggingface.co/Pygmalion/Pygmalion-6b

By their respective authors.

Warning: PPO_Pygway-6b may generate NSFW or inappropriate content due to the base models (Mainly Pygmalion/Pygmalion-6b) being trained on general user logs, and internet archives.

Intended Use:

Research purposes only, intended for responsible use. Express a conversation in natural language, and PPO_Pygmalion will pick up on the conversational format. Try starting a two line prompt such as:

Bot: "Hello, how are you?"
You: "I am doing just fine, thank you."

Or any other topic, and the model will carry on in this back and forth style.

Information:

For more details, check out the related source models, especially Pygmalion/Pygmalion-6b for more information on how to utilize the chat bot formatting expected.

In a similar manner to fine-tuning, merging weights does not add information but transforms it, therefore it is important to consider trade-offs. PPO_Pygway combines ppo_hh_gpt-j, Janeway-6b and Pygmalion-6b; all three models were blended in a two step process using a simple weighted parameter method

(X*A + Y*B)

With X & Y being the model weighs, and A/B being how strongly they are represented within the final value. The intent of this is to elevate the end-model by borrowing the strongly represented aspects out of each base model, but may also weaken other faces of each model, which can be desirable if the base models have problematic traits that need to be worked on.

Blend was done in FP32 and output saved in FP16 for reduced storage needs.

Limitations and biases

Based on known problems with NLP technology, potential relevant factors include bias (gender, profession, race and religion).

<ins>Warning: This model has a moderate NSFW bias.</ins>

License

BibTeX entry and citation info

@misc{gpt-j,
  author = {Wang, Ben and Komatsuzaki, Aran},
  title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}},
  howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
  year = 2021,
  month = May
}

Credits To:

Models involved:

Average weights merging Script credit to Concedo:

https://huggingface.co/concedo