© 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. 128444
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric rating values (e.g. “93.5”, “80.5”, “300”) in review-style contexts.

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

    by, ratings, how, gambling, use, Adele, 1

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     dolom
    -1.02
    JsonInclude
    -1.01
    czeniu
    -0.97
    чя
    -0.94
    Klub
    -0.92
    舂
    -0.92
    STRACT
    -0.91
    Kdo
    -0.91
    écial
    -0.91
    毫米
    -0.90
    POSITIVE LOGITS
     app
    1.23
     this
    1.23
    刚
    1.09
    دا
    1.06
     детские
    1.02
     isn
    0.99
    isuke
    0.97
     детской
    0.96
     hoang
    0.96
     pautas
    0.94
    Activations Density 0.003%

    No Known Activations