Massive MoE models ≥100B quantized with HLWQ · consumer deploy via vLLM expert offload
caio vicentino PRO
caiovicentino1
AI & ML interests
None yet
Recent Activity
updated a dataset about 6 hours ago
caiovicentino1/openinterp-43-multiprobe-grpo-full updated a dataset about 8 hours ago
caiovicentino1/openinterp-47-probe-gated-memory published a dataset about 8 hours ago
caiovicentino1/openinterp-47-probe-gated-memoryOrganizations
None yet
HLWQ Models
Hadamard-Lloyd Weight Quantization · arXiv:2603.29078 · formerly PolarQuant
-
caiovicentino1/Qwen3.5-9B-HLWQ-Q5
Text Generation • 9B • Updated • 1.94k • 3 -
caiovicentino1/Qwen3.5-9B-HLWQ-MLX-4bit
Text Generation • 1B • Updated • 2.96k • 3 -
caiovicentino1/Qwen3.5-27B-HLWQ-Q5
Text Generation • 27B • Updated • 4.41k • 10 -
caiovicentino1/Qwen3.5-9B-HLWQ-Engine-v4
Text Generation • 7B • Updated • 522
HLWQ Large MoE (100B+)
Massive MoE models ≥100B quantized with HLWQ · consumer deploy via vLLM expert offload
HLWQ Models
Hadamard-Lloyd Weight Quantization · arXiv:2603.29078 · formerly PolarQuant
-
caiovicentino1/Qwen3.5-9B-HLWQ-Q5
Text Generation • 9B • Updated • 1.94k • 3 -
caiovicentino1/Qwen3.5-9B-HLWQ-MLX-4bit
Text Generation • 1B • Updated • 2.96k • 3 -
caiovicentino1/Qwen3.5-27B-HLWQ-Q5
Text Generation • 27B • Updated • 4.41k • 10 -
caiovicentino1/Qwen3.5-9B-HLWQ-Engine-v4
Text Generation • 7B • Updated • 522
spaces 15
pinned
Sleeping
Agents
FabricationGuard Live Demo
🛡
Real-time fabrication detection on Qwen3.6-27B
pinned
Running
Agents
Qwen3.6 SAE Demo
🔬
Live token-level SAE feature for Qwen3.6-27B (AUROC 0.84)
pinned
Configuration error
OpenInterp
🔬
Watch language models think. Open source interpretability.
pinned
Paused
Agents
PolarQuant OmniWeaving Video
🧊
pinned
Paused
Agents
PolarQuant Demo
🧊
Configuration error
Agents
Qwen3.5-9B-Neo PolarQuant
🧊
models 64
caiovicentino1/Qwen3.6-35B-A3B-SAE-L23-topk-wip
Updated
caiovicentino1/qwen3.5-4b-crosscoder-rl-diff-papergrade
Updated
caiovicentino1/gemma2-2b-crosscoder-model-diff-papergrade
Updated
caiovicentino1/qwen36-27b-sae-papergrade
Updated • 4
caiovicentino1/qwen36-27b-sae-multilayer
Text Generation • Updated
caiovicentino1/qwen36-feature-circuits
Updated
caiovicentino1/qwen36-crest-cognitive-heads
Updated
caiovicentino1/qwen35-a3b-sae-phase2
Updated
caiovicentino1/Huihui-Qwopus3.5-27B-v3-abliterated-HLWQ-Q5
Text Generation • 26B • Updated • 3.12k • 14
caiovicentino1/Qwen3.5-4B-SAE-L18-topk
Feature Extraction • Updated • 1
datasets 19
caiovicentino1/openinterp-43-multiprobe-grpo-full
Updated
caiovicentino1/openinterp-47-probe-gated-memory
Preview • Updated
caiovicentino1/openinterp-46-cross-distribution-ensemble
Updated • 2
caiovicentino1/openinterp-45-inference-ensemble
Viewer • Updated • 6 • 24 • 1
caiovicentino1/openinterp-44-behavior-eval
Updated • 2
caiovicentino1/openinterp-42-multiprobe-grpo-pilot
Updated • 3
caiovicentino1/openinterp-41v2-grokking-extended
Viewer • Updated • 2 • 3
caiovicentino1/openinterp-37v2-multiprobe-dpo-extended
Preview • Updated • 3
caiovicentino1/openinterp-41-grokking-forward-only
Viewer • Updated • 2 • 4
caiovicentino1/openinterp-37-multiprobe-dpo-full
Updated • 23