llm.mojo GPT-2 releases Model checkpoints trained from scratch with llm.mojo (Mojo/MAX port of Karpathy's llm.c). https://github.com/ulmentflam/llm.mojo ulmentflam/gpt2-124m-fineweb-mojo Text Generation • 0.1B • Updated Jul 27 • 52 ulmentflam/gpt2-774m-fineweb-mojo Text Generation • 0.8B • Updated Jul 30 • 23 ulmentflam/gpt2-124m-fineweb-fp8-mojo Text Generation • 0.1B • Updated Jul 28 • 17 ulmentflam/gpt2-774m-fineweb-fp8-mojo Text Generation • 0.8B • Updated Jul 31 • 21
llm.mojo GPT-2 releases Model checkpoints trained from scratch with llm.mojo (Mojo/MAX port of Karpathy's llm.c). https://github.com/ulmentflam/llm.mojo ulmentflam/gpt2-124m-fineweb-mojo Text Generation • 0.1B • Updated Jul 27 • 52 ulmentflam/gpt2-774m-fineweb-mojo Text Generation • 0.8B • Updated Jul 30 • 23 ulmentflam/gpt2-124m-fineweb-fp8-mojo Text Generation • 0.1B • Updated Jul 28 • 17 ulmentflam/gpt2-774m-fineweb-fp8-mojo Text Generation • 0.8B • Updated Jul 31 • 21