Add detailed report for Qwen2.5-0.5B ↔ Gemma-2-2B-it experiment d4e3c03 verified Samarth0710 commited on 29 days ago
REPORT_REASONING.md v2: retract oracle-beating claim, document strict vs lenient eval, weak v1 oracle recipe 967b676 verified Samarth0710 commited on Apr 24
reasoning v2: stronger oracle recipe, strict+lenient eval, refit mappings 8ddc968 verified Samarth0710 commited on Apr 24
Add cross-model LoRA adapter prediction — reasoning validation (GSM8K) 4a74ac9 verified Samarth0710 commited on Apr 23
Upload scaled/results_phaseB.json with huggingface_hub a577216 verified Samarth0710 commited on Apr 22
Upload scaled/results_phaseA.json with huggingface_hub ac2fc21 verified Samarth0710 commited on Apr 22
Add scaled experiment: 25 anchors, 5 held-out tasks, 5 mapping methods c9d1381 verified Samarth0710 commited on Apr 22