© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 22326
    Prev
    Next
    INDEX
    Explanations

    programming and errors

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

    The neuron fires most strongly on unusually long tokens (e.g. long method or class names, long words), effectively detecting token length.

    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
    s
    -2.92
    </td>
    -2.66
     voor
    -2.48
    猡
    -2.25
    }$
    -2.14
    line
    -2.14
     необходимо
    -2.14
     régionales
    -2.14
    h
    -2.13
     coloriage
    -2.09
    POSITIVE LOGITS
    ↵
    2.84
    …”
    2.72
    !”
    2.58
    0
    2.31
    我也不
    2.28
    骜
    2.23
     ältere
    2.22
    誂
    2.22
     kostenlose
    2.13
     dão
    2.13
    Activations Density 0.010%

    No Known Activations