VOOZH about

URL: https://huggingface.co/xiaol/RWKV-4-world-one-state-ultrachat-COT-65k

โ‡ฑ xiaol/RWKV-4-world-one-state-ultrachat-COT-65k ยท Hugging Face


RWKV v4 7B world model

finetuned with ultrachat , COT and some novel instructions data, commitpackft and so on

use full ultrachat and cot data, about 3B tokens

if you wanna do Role play, use this model

Contributor

@JL-er @Remixa

Design of experiment

this model lose multi-turn chat ability,cause from using whole ultrachat datasets.

so i continue tuned multi-turn datasets with 2 aspects

1.role play

2.novel multiturn instruction

Training details

wandb.ai

CAses

๐Ÿ‘ image/jpeg

๐Ÿ‘ image/jpeg

Usage

adjust tempp and topp on different scenario. ๐Ÿ‘ image/png

๐Ÿ‘ image/png

๐Ÿ‘ image/png

COT and lookback

๐Ÿ‘ image/png
this model can do above task with 100% acc.

๐Ÿ‘ image/png

role play model

๐Ÿ‘ image/png

๐Ÿ‘ image/png

novel

๐Ÿ‘ image/png

๐Ÿ‘ image/png

demo site(temporary)

online showcase

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Datasets used to train xiaol/RWKV-4-world-one-state-ultrachat-COT-65k

Collection including xiaol/RWKV-4-world-one-state-ultrachat-COT-65k