🔄 In a Training Loop

SSH PRO

ProCreations

·

https://ssh.codes

AI & ML interests

Making AI’s small, accessible, and easy to run on consumer hardware

Recent Activity

repliedto DavidAU's post about 15 hours ago

Qwen3.6-40B-Fable-Fusion-6-Core-Deckard-Eleanor-Heretic-Uncensored The fire of multiple Fable Fusion 711 cores in one, larger model. This is the first 40B fine tune that reaches "closed source" (IE OpenAI, Claude) level of intelligence in both 8 bit and 4 bit. This model is composed from multiple Qwen 27B Fable Fusion 711 cores (1700+ likes, 2.3 million+ downloads) - a record breaking model in terms of intelligence and raw power. The "40B Eleanor" takes this to the next level with improvements in thinking tokens/ thinking block size (1/10 to 1/2 the size), thinking in general and output detail quality with deep analytics too. - 1/10 to 1/2 the number of thinking tokens. - Extreme depth of detail in generations, including long form, in depth analytics. - STRONG creative abilities. - Auto-variable reasoning: Model only reasons as much as the task requires. - Strong general intelligence. - It says what it means in less words, more clearly than any previous tuned model. - It will go all in, in exacting detail when the situation calls for it. - If it thinks something is wrong / wrong path it will say so too. Regular and MTP Quants: https://huggingface.co/DavidAU/Qwen3.6-40B-Fable-Fusion-6-Core-Deckard-Eleanor-Heretic-Uncensored-NM-DAU-NEO-MAX-MTP-GGUF

updated a Space 1 day ago

ProCreations/maple-webgpu

liked a Space 1 day ago

ProCreations/maple-webgpu

View all activity

Organizations

ProCreations 's Spaces 354

Reproduction: Statistical Evaluability of Generative Models

Explore a research logbook and connect with a coding agent

Reproduction: Improved Convergence Analysis of Topology Dependence in Decentralized SGD

Browse and collaborate on research logbooks

Reproduction: High-Probability Convergence Guarantees of Decentralized SGD

Explore a research logbook and sync with an AI agent

Reproduction: Mirror Descent Under Generalized Smoothness

Explore and collaborate on research logbooks

Reproduction: Box Thirding: Anytime Best Arm Identification under Insufficient Sampling

Track research findings and sync with an AI coding agent

Reproduction: Towards Understanding Adam Convergence on Highly Degenerate Polynomials

Explore research logbooks and collaborate with an AI agent

Reproduction: Catastrophic k-Means Failure and Hartigan's Escape

Explore research logbooks and collaborate with an AI agent

Reproduction: f-Trajectory Balance

Explore a collaborative logbook for your coding agent

Reproduction: Can Adaptive Gradient Methods Converge under Heavy-Tailed Noise? A Case Study of AdaGrad

Browse and sync logbook entries with an AI coding agent

Reproduction - A Tight Theory of Error Feedback Algorithms in Distributed Optimization

Exact EF rates, cycles, cubic and EControl audit

Reproduction - Masked SSL Random Matrix Theory

BBP theory and a spiked-PCA falsification

Reproduction - Microcanonical Mini-Batch Noise

Paper-scale Fashion-MNIST BNN noise and energy audit

Reproduction - Heavy-Tailed Data and Gradient Clipping

Explore and share experiment logs with an AI collaborator

Repro - Approximate Proportionality Online

View and collaborate on audit logbooks with an AI agent

Repro - Learning for Branch-and-Cut

Collaborate on a logbook and sync findings with your coding agent

Repro - Row-Stochastic Matrices Can Provably Outperform Doubly Stochastic Matrices

Explore and collaborate on research logbooks online

Flat Minima and Generalization — Exact SCO Reproduction

Browse and share your project logbook with an AI agent

Softmax as Linear Attention — Large-Prompt Gaussian Reproduction

Browse and sync your logbook with an AI agent