All 8 seeds of d16 (268M params), 200 TPP on ClimbMix; plain GPT (all optional nanochat mechanisms disabled). Anneal snapshots as revisions.
Julian Minder
jkminder
AI & ML interests
None yet
Recent Activity
published a model about 2 hours ago
jkminder/d16_268m_seed8 updated a collection about 2 hours ago
Scaling Ladder — 4 sizes, 135M–973M, 200 TPP updated a collection about 2 hours ago
d16 — 268M params, 8 seeds, 200 TPPOrganizations
models 140
jkminder/d16_268m_seed8
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed7
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed6
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed5
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed4
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed2
Text Generation • 0.3B • Updated
jkminder/d16_268m_seed3
Text Generation • 0.3B • Updated
jkminder/d26_973m_seed4
Text Generation • 1.0B • Updated
jkminder/d26_973m_seed3
Text Generation • 1.0B • Updated
jkminder/d26_973m_seed2
Text Generation • 1.0B • Updated
datasets 15
jkminder/pretraining-priors-pirate-register
Viewer • Updated • 999k • 115
jkminder/normative-reflections-feedback
Viewer • Updated • 5 • 10
jkminder/apertus-annotation-feedback
Viewer • Updated • 52 • 16
jkminder/model-raising-interp-prepared
Viewer • Updated • 100k • 29
jkminder/model-raising-pbsft-eval
Viewer • Updated • 9.99k • 26
jkminder/model-raising-reflection-end-eval
Viewer • Updated • 28.7k • 13
jkminder/model-raising-pb-300k-3c-sft
Viewer • Updated • 302k • 14
jkminder/or-bench-1k-overrefusal
Viewer • Updated • 962 • 34
jkminder/xstest-overrefusal
Viewer • Updated • 214 • 40
jkminder/orfuzz
Viewer • Updated • 1.79k • 9