227 132 2697

Knut Jägersberg

KnutJaegersberg

jagersbergknut

AI & ML interests

NLP, opinion mining, narrative intelligence

Recent Activity

reacted to BlinkDL's post with 🔥 about 5 hours ago

RWKV-7 "Goose" 0.4B trained w/ ctx4k automatically extrapolates to ctx32k+, and perfectly solves NIAH ctx16k 🤯 100% RNN and attention-free. Only trained on the Pile. No finetuning. Replicable training runs. tested by our community: https://github.com/Jellyfish042/LongMamba

liked a model about 6 hours ago

google/siglip2-giant-opt-patch16-384

upvoted a collection about 6 hours ago

SigLIP2

View all activity

Organizations

KnutJaegersberg's activity

reacted to BlinkDL's post with 🔥 about 5 hours ago

Post

3779

liked a model about 6 hours ago

google/siglip2-giant-opt-patch16-384

Zero-Shot Image Classification • Updated about 12 hours ago • 10 • 4

upvoted a collection about 6 hours ago

SigLIP2

Collection

36 items • Updated about 7 hours ago • 27

liked a model about 11 hours ago

PrimeIntellect/SYNTHETIC-1-SFT-7B

Updated about 14 hours ago • 82 • 4

upvoted a collection about 11 hours ago

SYNTHETIC-1

Collection

A collection of tasks & verifiers for reasoning datasets • 9 items • Updated about 17 hours ago • 43

liked a model about 22 hours ago

HuggingFaceTB/SmolVLM2-256M-Video-Instruct

Video-Text-to-Text • Updated 1 day ago • 204 • 16

commented on Mimicking Consciousness in LLMs: Ascending the Dimensions of Thought with Recurrent Processing 1 day ago

also more detail in the AI podcasts:

https://www.youtube.com/watch?v=OcI6pMBcEbs

https://www.youtube.com/watch?v=elkGpl3sTig

posted an update 1 day ago

Post

525

Mimicking Consciousness in LLMs: Ascending the Dimensions of Thought with Recurrent Processing

This blog post explores how **recurrent processing** can transform Large Language Models (LLMs) to mimic aspects of human thought by engaging in iterative feedback loops. Inspired by string theory, the post describes how LLMs can "ascend dimensions" of cognition, progressing through foundational cognitive loops—such as basic cognition, executive functions, and meta-cognition—before advancing into **world simulation**. In this stage, LLMs explore higher dimensions, perceiving non-linear time, simulating branching possibilities, and integrating multiple realities. The interaction between the **Generator** and **Reflective Compass** allows AI systems to refine their outputs iteratively, moving toward a **point attractor** where ideas become coherent and polished. While this process doesn't bestow true consciousness, it offers a compelling imitation of reflective and adaptive thinking, leading to smarter dialogue, enhanced creativity, and more robust problem-solving.

https://huggingface.co/blog/KnutJaegersberg/oscillatory-recurrence-for-llms