© 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. 63938
    Prev
    Next
    INDEX
    Explanations

    gaming roles and performance

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

    The neuron spots domain‐specific proper names and technical terms (e.g. game characters, team names, APIs, 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_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
     stuffy
    -0.75
     kroppen
    -0.73
    =_
    -0.70
    tinger
    -0.69
    EMBLY
    -0.69
     Wab
    -0.68
     Chama
    -0.68
    érrez
    -0.68
     Yr
    -0.66
    𝒾
    -0.65
    POSITIVE LOGITS
     Support
    1.30
     support
    1.26
    support
    1.17
    Support
    1.16
     carry
    1.04
     SUPPORT
    1.04
    Supports
    1.03
    SUPPORT
    1.01
     Tank
    1.00
     tank
    0.98
    Activations Density 0.015%

    No Known Activations