Hugging Face's logo
VIDRAFT_LAB's picture
In a Training Loop 🔄
11 9 74

VIDRAFT_LAB

SeaWolf-AI
omiddhn's profile picture Rickie77's profile picture Anserwise's profile picture
·

AI & ML interests

None yet

Recent Activity

repliedto their post 17 minutes ago
🧬 Darwin V6: Diagnostic-Guided Evolutionary Model Merging We are releasing Darwin-31B-Opus — a reasoning-enhanced model merging Google's Gemma-4-31B-it and TeichAI's Claude Opus Distill using the Darwin V6 engine. Model: https://huggingface.co/FINAL-Bench/Darwin-31B-Opus Demo: https://huggingface.co/spaces/FINAL-Bench/Darwin-31B-Opus 🔬 What Darwin V6 Does Conventional merging tools (mergekit, etc.) apply a single ratio to all tensors. Set ratio=0.5 and all 1,188 tensors blend identically, with no distinction between which tensors matter for reasoning versus coding. Darwin V6 diagnoses both parents at the tensor level before merging. It measures Shannon entropy, standard deviation, and L2 norm for every tensor, then passes 5 diagnostic probes (REASONING, CODE, MATH, KNOWLEDGE, LANGUAGE) through the model to determine layer-wise functional importance. Each of the 1,188 tensors receives an independent optimal ratio. combined = static(entropy/std/norm) x 0.4 + probe(cosine_distance) x 0.6 final_ratio = mri_ratio x mri_trust + genome_ratio x (1 - mri_trust) When one parent is overwhelmingly superior for a tensor (ratio < 0.15 or > 0.85), Darwin transplants it directly without interpolation. The mri_trust parameter itself is optimized by CMA-ES evolutionary search, so optimal transplant intensity is determined automatically. After merging, a Health Check compares the child against both parents layer-by-layer to detect interference or function loss. 🧬 Parent Models Father: google/gemma-4-31B-it Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill 🧬 Results Compared under identical conditions (same 50 questions, same seed, greedy, thinking mode): Father: 60.0% (30/50) Darwin-31B-Opus: 66.0% (33/50) — +10% relative improvement ARC-Challenge: 82.89% (loglikelihood, zero-shot, 200 questions) Optimal genome found by evolution: ffn_ratio=0.93 — FFN layers strongly favor Mother (Claude Opus Distill) block_5 (L50-L59)=0.86 and more...
repliedto their post 21 minutes ago
🧬 Darwin V6: Diagnostic-Guided Evolutionary Model Merging We are releasing Darwin-31B-Opus — a reasoning-enhanced model merging Google's Gemma-4-31B-it and TeichAI's Claude Opus Distill using the Darwin V6 engine. Model: https://huggingface.co/FINAL-Bench/Darwin-31B-Opus Demo: https://huggingface.co/spaces/FINAL-Bench/Darwin-31B-Opus 🔬 What Darwin V6 Does Conventional merging tools (mergekit, etc.) apply a single ratio to all tensors. Set ratio=0.5 and all 1,188 tensors blend identically, with no distinction between which tensors matter for reasoning versus coding. Darwin V6 diagnoses both parents at the tensor level before merging. It measures Shannon entropy, standard deviation, and L2 norm for every tensor, then passes 5 diagnostic probes (REASONING, CODE, MATH, KNOWLEDGE, LANGUAGE) through the model to determine layer-wise functional importance. Each of the 1,188 tensors receives an independent optimal ratio. combined = static(entropy/std/norm) x 0.4 + probe(cosine_distance) x 0.6 final_ratio = mri_ratio x mri_trust + genome_ratio x (1 - mri_trust) When one parent is overwhelmingly superior for a tensor (ratio < 0.15 or > 0.85), Darwin transplants it directly without interpolation. The mri_trust parameter itself is optimized by CMA-ES evolutionary search, so optimal transplant intensity is determined automatically. After merging, a Health Check compares the child against both parents layer-by-layer to detect interference or function loss. 🧬 Parent Models Father: google/gemma-4-31B-it Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill 🧬 Results Compared under identical conditions (same 50 questions, same seed, greedy, thinking mode): Father: 60.0% (30/50) Darwin-31B-Opus: 66.0% (33/50) — +10% relative improvement ARC-Challenge: 82.89% (loglikelihood, zero-shot, 200 questions) Optimal genome found by evolution: ffn_ratio=0.93 — FFN layers strongly favor Mother (Claude Opus Distill) block_5 (L50-L59)=0.86 and more...
reacted to theirpost with 👀 about 2 hours ago
🧬 Darwin V6: Diagnostic-Guided Evolutionary Model Merging We are releasing Darwin-31B-Opus — a reasoning-enhanced model merging Google's Gemma-4-31B-it and TeichAI's Claude Opus Distill using the Darwin V6 engine. Model: https://huggingface.co/FINAL-Bench/Darwin-31B-Opus Demo: https://huggingface.co/spaces/FINAL-Bench/Darwin-31B-Opus 🔬 What Darwin V6 Does Conventional merging tools (mergekit, etc.) apply a single ratio to all tensors. Set ratio=0.5 and all 1,188 tensors blend identically, with no distinction between which tensors matter for reasoning versus coding. Darwin V6 diagnoses both parents at the tensor level before merging. It measures Shannon entropy, standard deviation, and L2 norm for every tensor, then passes 5 diagnostic probes (REASONING, CODE, MATH, KNOWLEDGE, LANGUAGE) through the model to determine layer-wise functional importance. Each of the 1,188 tensors receives an independent optimal ratio. combined = static(entropy/std/norm) x 0.4 + probe(cosine_distance) x 0.6 final_ratio = mri_ratio x mri_trust + genome_ratio x (1 - mri_trust) When one parent is overwhelmingly superior for a tensor (ratio < 0.15 or > 0.85), Darwin transplants it directly without interpolation. The mri_trust parameter itself is optimized by CMA-ES evolutionary search, so optimal transplant intensity is determined automatically. After merging, a Health Check compares the child against both parents layer-by-layer to detect interference or function loss. 🧬 Parent Models Father: google/gemma-4-31B-it Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill 🧬 Results Compared under identical conditions (same 50 questions, same seed, greedy, thinking mode): Father: 60.0% (30/50) Darwin-31B-Opus: 66.0% (33/50) — +10% relative improvement ARC-Challenge: 82.89% (loglikelihood, zero-shot, 200 questions) Optimal genome found by evolution: ffn_ratio=0.93 — FFN layers strongly favor Mother (Claude Opus Distill) block_5 (L50-L59)=0.86 and more...
View all activity

Organizations

FINAL_Bench's profile picture