This is a GPT-2 (1558M) model trained in llm.c for 100B tokens with cosine LR on DCLM-Baseline.

A lot more detailed info and observations are here: https://x.com/Yuchenj_UW/status/1813260100192334108

Downloads last month
132
Safetensors
Model size
1.56B params
Tensor type
BF16
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Dataset used to train yuchenj/gpt2_1558M_100B_DCLM_hf