© 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. 78855
    Prev
    Next
    INDEX
    Explanations

    free software strings

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

    The neuron fires on code‐style or technical tokens—especially numeric literals, underscores, and other non-alphabetic symbols.

    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
     MDC
    -0.73
     Botanic
    -0.71
     WK
    -0.70
     CMD
    -0.69
     няма
    -0.68
     HDR
    -0.66
     wilson
    -0.66
    Thr
    -0.65
    ANCED
    -0.65
     mama
    -0.65
    POSITIVE LOGITS
    عط
    0.79
     suite
    0.73
     freezes
    0.71
     đèn
    0.69
     Сан
    0.69
    masking
    0.68
     Saúde
    0.66
     MainAxisSize
    0.66
    خی
    0.65
    erta
    0.65
    Activations Density 0.074%

    No Known Activations