A failed experiment. Lowered v10's weight decay from .005 to .001.

Interested? see RLLM Virtual map for more context.

Downloads last month
-
Safetensors
Model size
2B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Space using migueldeguzmandev/GPT2XL_RLLMv10-wd-001 1

Collection including migueldeguzmandev/GPT2XL_RLLMv10-wd-001