CleanGPT

This is a clean model based on the GPT-2 small architecture, this model does not have training, it is an untrained model.

why so?

A model with this form is a ready-made model that we can use at any time to train and work on it and not on GPT-2, which may be limited to its old training data, making it impossible to extract its greatest performance.

Downloads last month
14
Safetensors
Model size
124M params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support