© 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. 128161
    Prev
    Next
    INDEX
    Explanations

    The neuron flags highly technical tokens—like gene/allele names, Greek‐letter variables, specialized mathematical terms, and numeric measurements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    genetic terms and equations

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

    mathematical or programming notation, particularly brackets, variables, and assignment operators.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    ITEN
    -0.93
     Besoin
    -0.92
     Toutes
    -0.88
     Todas
    -0.88
     budow
    -0.88
     chefe
    -0.87
     Información
    -0.83
     Organisateur
    -0.81
    Percy
    -0.80
     нашу
    -0.80
    POSITIVE LOGITS
    Either
    0.89
    鴣
    0.86
     comprehend
    0.86
     when
    0.85
    CURR
    0.85
    ــ
    0.84
    accéder
    0.84
    oolean
    0.84
    お待ちしております
    0.84
     comprendere
    0.83
    Activations Density 0.026%

    No Known Activations