Downloads · 30 days
21
33% of all-time downloads
AmeerH/FPT-Base
FPT-Base is a text generation model from AmeerH. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as mit.
This is a GPT-2 model trained for 330K steps from scratch (of 1M batch size) on FineWeb-EDU i.e around 300B Tokens.
Downloads · 30 days
21
33% of all-time downloads
All-time downloads
64
Public
Parameters
1.6B
3.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors3.1 GB · 100%
From the Hugging Face model README
This is a GPT-2 model trained for 330K steps from scratch (of 1M batch size) on FineWeb-EDU i.e around 300B Tokens.
This is the model card of a 🤗 transformers model that has been pushed on the Hub..
Developed by: Ameer H Shared by [optional]: Ameer H Model type: GPT2 Language(s) (NLP): English License: MIT
Will produce blabbers and unintended slurs racial or anything. Do not blame this is just an experiment.
Forked from Andrej Karparthy's original model.