Onlydrinkwater's picture
Upload README.md with huggingface_hub
57b6325 verified
metadata
tags:
  - coconut
  - gpt2
  - reasoning
base_model: openai-community/gpt2

gpt2-coconut-checkpoint4

CoCoNut (Chain of Continuous Thought) trained checkpoint 4 based on GPT-2.

Training

  • Base model: openai-community/gpt2
  • Training method: CoCoNut (continuous latent reasoning)
  • Dataset: GSM8K

Usage

import torch
checkpoint = torch.load("pytorch_model.bin")