© 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. 34-GEMMASCOPE-RES-131K
    7. 31645
    Prev
    Next
    INDEX
    Explanations

    long followed by other words

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

    This neuron activates on uses of “long” (including hyphenated or extended forms like “long-ish” or “long-distance”), signaling mentions of extended duration or 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_34/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
    fins
    -0.75
     徒歩
    -0.73
     postnatal
    -0.72
    єр
    -0.71
    asc
    -0.70
    แน
    -0.69
    れい
    -0.69
     unpack
    -0.69
    Kondisi
    -0.69
     Khal
    -0.69
    POSITIVE LOGITS
     sized
    0.91
    SizedBox
    0.82
    sized
    0.76
     size
    0.76
     long
    0.75
     PROCEDURES
    0.75
     SIZE
    0.73
    ifolia
    0.73
     スカート
    0.73
     Hours
    0.71
    Activations Density 0.043%

    No Known Activations