© 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. 22-GEMMASCOPE-RES-131K
    7. 15645
    Prev
    Next
    INDEX
    Explanations

    numbers and ranges

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

    The neuron is effectively silent on all shown tokens and does not detect any particular pattern (i.e. it appears inactive).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     trám
    -1.24
    avaa
    -1.16
     também
    -1.15
    Apparently
    -1.10
     atencion
    -1.06
    Either
    -1.06
     secado
    -1.05
     dinosaurios
    -1.05
    apparently
    -1.05
    either
    -1.05
    POSITIVE LOGITS
    出版年
    0.96
    Francia
    0.94
    intă
    0.93
     २०
    0.90
    Greta
    0.90
    ㌧
    0.89
     ndani
    0.87
     landes
    0.85
    ykite
    0.84
    Blazor
    0.84
    Activations Density 0.021%

    No Known Activations