miguelcsx/causal-focus-bbpe16k-tokenizer
Updated
CausalFocus GPT-BERT D256
BabyLM Strict-Small release artifacts for CausalFocus-GPTBERT-D256-BBPE16K and the Objective Sets The Rule study.
Note tokenizer
miguelcsx/babylm-bbpe16k-512-encoded
Updated • 69Note encoded corpus
miguelcsx/babylm-bbpe16k-512-structured-priors
Updated • 101Note structured priors