OJ

ojaffe 's models 198

Sort:  Recently updated

ojaffe/20260413-091816-babylm-5f3c-2026-04-15-033-pure-kl-topk500

0.2B • Updated Apr 15 • 2

ojaffe/2026-04-15-029-lr4e3-swa

0.2B • Updated Apr 15 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-15-032-pure-kl-topk200

0.2B • Updated Apr 15 • 3

ojaffe/2026-04-15-028-very-high-lr-swa

0.2B • Updated Apr 15 • 4

ojaffe/20260413-091816-babylm-5f3c-2026-04-15-031-pure-kl-t1-large

0.2B • Updated Apr 15 • 2

ojaffe/2026-04-15-027-high-lr-swa

0.2B • Updated Apr 15 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-030-pure-kl-t1-wider

0.2B • Updated Apr 15 • 3

ojaffe/2026-04-14-026-no-dropout-swa

0.2B • Updated Apr 15 • 2

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-029-pure-kl-t1-constant

0.1B • Updated Apr 14 • 3

ojaffe/2026-04-14-025-low-dropout-swa

0.2B • Updated Apr 14 • 4

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-028-pure-kl-fast-anneal

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-027-pure-kl-const-temp

0.1B • Updated Apr 14 • 1

ojaffe/world-model

Updated Apr 14 • 1

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-026-pure-kl-topk50

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-025-pure-kl-topk

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-024-fast-anneal-high-lr

0.1B • Updated Apr 14 • 4

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-023-wide-deep-topk-anneal

0.2B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-022-deeper-topk-anneal

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-021-topk-temp-anneal-deep

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-020-deep-narrow-gpt2

0.1B • Updated Apr 14 • 4

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-019b-custom-tokenizer-100k

0.1B • Updated Apr 14 • 4

ojaffe/20260413-211548-small-world-model-633d-2026-04-14-013-2step-rollout

Updated Apr 14 • 2

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-018-opt6.7b-teacher

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-017-feature-distillation

0.1B • Updated Apr 14 • 3

ojaffe/20260413-091816-babylm-5f3c-2026-04-14-016-distill-llama3-3b

0.2B • Updated Apr 14 • 2

ojaffe/2026-04-14-044-xl-wide

0.5B • Updated Apr 14 • 3

ojaffe/2026-04-13-043-wider-student

0.5B • Updated Apr 14 • 3

ojaffe/2026-04-13-042-large-student

0.4B • Updated Apr 13 • 3

ojaffe/2026-04-13-064-tqa-dpo-on-062

0.6B • Updated Apr 13 • 2

ojaffe/2026-04-13-041-rdrop2

0.2B • Updated Apr 13 • 3