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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 42331
    Prev
    Next
    INDEX
    Explanations

    stuttering and tic disorders

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on tokens naming or describing speech disfluencies—e.g. stuttering, stumbling, slips or hesitations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     There
    -1.80
     If
    -1.77
     escrito
    -1.53
    Some
    -1.52
     has
    -1.50
    良好的
    -1.47
     Many
    -1.47
    Year
    -1.45
     Their
    -1.45
    emoc
    -1.43
    POSITIVE LOGITS
    ',
    2.16
    ”.
    2.00
    ).
    1.89
    })$.
    1.82
    }
    1.77
    錒
    1.76
    ᯅ
    1.70
     tinham
    1.67
    }`).
    1.66
    }{
    1.65
    Activations Density 0.010%

    No Known Activations