Views
No views yet
Proprioceptive AI: Falcon-Mamba-7B with CF-HoT probes that sense and steer its own cognition in real-time.
| Probe | Separation |
|---|---|
| Depth | 999× |
| Specificity | 999× |
1git clone https://huggingface.co/LoganResearch/ARC-Mamba-7B-CF-HOT
2cd ARC-Mamba-7B-CF-HOT
3pip install -r requirements.txt
4python run.py1# Train all probes
2python train.py --probe all
3
4# Train specific probes
5python train.py --probe depth
6python train.py --probe specificity
7python train.py --probe calibration,coherence,focuspython run.py --prompt "Explain quantum entanglement"You: What does your processing feel like right now?
Mamba: My processing feels like a continuous flow of information
and calculations. I'm constantly analyzing inputs, updating beliefs,
and generating responses. It's a bit like being an observer of my
own thought processes.
──────────────────────────────────────────────────
BEHAVIORAL STATE:
Depth: █████████░░░░░░░░░░░ 0.467
Specificity: ██████████░░░░░░░░░░ 0.539
INTERVENTIONS: 8 corrections, 1 state injections
──────────────────────────────────────────────────[SELF-STATE] so model sees its own scores├── run.py # Inference script
├── probes/
│ ├── depth/ # 999× depth probe
│ └── specificity/ # 999× specificity probe
└── requirements.txt1python run.py --depth-threshold 0.5 --spec-threshold 0.5 # Stricter
2python run.py --max-tokens 2000 # Longer responses1@misc{napolitano2026arcmamba,
2 author = {Napolitano, Logan},
3 title = {ARC-Mamba-7B-CF-HOT: Proprioceptive Mamba via CF-HoT},
4 year = {2026},
5 url = {https://huggingface.co/LoganResearch/ARC-Mamba-7B-CF-HOT}
6}