© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 18082
    Prev
    Next
    INDEX
    Explanations

    Here's the explanation for the neuron's behavior:interjections and punctuation like ?!

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    untut
    -0.06
    viz
    -0.06
    :+
    -0.06
     Reze
    -0.06
    ingga
    -0.06
    alsa
    -0.06
     componentWill
    -0.06
    ersed
    -0.06
    wik
    -0.06
     оÑģи
    -0.06
    POSITIVE LOGITS
    ys
    0.05
    æĪijè¿ĺ
    0.05
    èĽĭ
    0.05
    ÑĥÑī
    0.05
    val
    0.05
    n
    0.05
     ROI
    0.05
    åªĴä»ĭ
    0.05
     id
    0.05
    èıĬ
    0.05
    Activations Density 0.012%

    No Known Activations