JThomas-CoE/coe-gemma4-biology-mmlu_pro-14b-a4b-q4

https://huggingface.co/JThomas-CoE/coe-gemma4-biology-mmlu_pro-14b-a4b-q4
Activeby JThomas-CoE901updated 3 weeks ago

Base model: google/gemma-4-26b-it Architecture: MoE — 26B total / ≈4B active parameters (1 shared expert + 8 routed from a pool of 128 per MoE layer, 30 MoE layers) Method: Activation-directed expert surgery — 128 → 64 experts per layer (50% reduction) Quantization: Q4KM (≈9.7 GB on disk) Tags:…

Sourced from

  • HuggingFaceJThomas-CoE/coe-gemma4-biology-mmlu_pro-14b-a4b-q4

Related resources

Base model: google/gemma-4-26b-it Architecture: MoE — 26B total / ≈4B active parameters (1 shared expert + 8 routed from a pool of 128 per MoE layer, 30 MoE layers) Method: Activation-directed expert surgery — 128 → 64 experts per layer (50% reduction) Quantization: Q4KM (≈9.7 GB on disk) Tags:…

Active3063 weeks ago

Abstract:

Stale8752 years ago
Python

From Inquiry to Decision: Building Trustworthy Medical AI

Active204 months ago
Python

Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants.

Idle9.5K1 year ago
Python

# or·a·cle /ˈôrəkəl/ — a source of wise counsel; one who provides authoritative knowledge. From Latin ōrāculum, meaning divine announcement. In computer science, an oracle is a black box that always returns the correct answer — you don't ask it how it knows, you ask and it answers.

Active1422 months ago
Python