© 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. 72490
    Prev
    Next
    INDEX
    Explanations

    climbing and ascent

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

    The neuron fires on specialized rock‐climbing terms (e.g. bouldering, trad, sport, climbing, crags, problems, moves, etc.).

    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
     different
    -1.44
     pushing
    -1.19
     here
    -1.19
    嗚
    -1.16
     delivering
    -1.16
    が変わ
    -1.16
    august
    -1.15
     looking
    -1.12
    bidi
    -1.12
     există
    -1.10
    POSITIVE LOGITS
    pakatan
    1.25
    ----</
    1.22
    THREE
    1.21
    ∮
    1.21
    nae
    1.20
     ванной
    1.20
     THREE
    1.18
     seguinte
    1.17
    QMetaType
    1.16
     vur
    1.16
    Activations Density 0.031%

    No Known Activations