Model Raising

AI & ML interests

None defined yet.

Recent Activity

Raghav-Singhal updated a Space about 7 hours ago

model-raising/README

Raghav-Singhal published a Space about 7 hours ago

model-raising/README

Raghav-Singhal updated a collection about 7 hours ago

View all activity

updated a Space about 7 hours ago

README

published a Space about 7 hours ago

README

updated 4 collections about 7 hours ago

SPP

Synthetic Persona Pretraining (SPP): 3B models from the paper. • 5 items • Updated about 7 hours ago

Pretraining Datasets

The SPP pretraining corpus: reflections, the corpus selection manifest, safety scores, and verification data. Fully reproducible from public Dolma 3. • 6 items • Updated about 7 hours ago

Post-training Dataset

SP-SFT: the post-training mix that binds the assistant identity to the persona installed during pretraining. • 2 items • Updated about 7 hours ago

Evals

ConstitutionEval (in-domain constitution following) and audited AIRiskDilemmas (out-of-distribution value prioritisation and risk-taking). • 2 items • Updated about 7 hours ago

updated a collection about 8 hours ago

Models - 1.7B

1.7B SPP models: bases publish their training trajectory as step-* revisions, plus instruct counterparts. • 10 items • Updated about 7 hours ago