Seed-twinned DataDecide models under one safety post-training recipe, with per-prompt judge labels.
AI & ML interests
Open and Efficient LLM Pretraining
Recent Activity
View all activity
models 31
cerulean-works/castor-dolma1_7-1B
Text Generation • Updated • 9
cerulean-works/castor-dclm-baseline-qc-10p-1B
Text Generation • Updated • 12
cerulean-works/castor-falcon-1B
Text Generation • Updated • 8 • 1
cerulean-works/castor-dclm-baseline-75p-dolma1.7-25p-1B
Text Generation • Updated • 8
cerulean-works/castor-falcon-and-cc-qc-orig-10p-1B
Text Generation • Updated • 7
cerulean-works/castor-dclm-baseline-50p-dolma1.7-50p-1B
Text Generation • Updated • 13
cerulean-works/castor-dolma1_7-no-code-1B
Text Generation • Updated • 9
cerulean-works/castor-falcon-and-cc-qc-tulu-10p-1B
Text Generation • Updated • 9
cerulean-works/castor-dclm-baseline-qc-fw-3p-1B
Text Generation • Updated • 8
cerulean-works/castor-dclm-baseline-qc-20p-1B
Text Generation • Updated • 10