© 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. 26700
    Prev
    Next
    INDEX
    Explanations

    Evolution package Phantom Q

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

    The neuron activates on standalone number tokens (e.g. line numbers, version numbers, and other numeric literals) in code or stack‐trace contexts.

    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
     couverts
    -1.36
     galeria
    -1.16
    loggers
    -1.12
    萏
    -1.11
     stockfoto
    -1.11
    rainian
    -1.09
    diode
    -1.09
     estimés
    -1.09
     plafon
    -1.09
    mois
    -1.09
    POSITIVE LOGITS
    Boxscore
    1.34
     and
    1.20
     renegade
    1.16
    这些人
    1.12
    肄
    1.06
     hydrochlor
    1.04
    锲
    1.03
     embers
    1.02
     anhyd
    1.02
    滿足
    1.01
    Activations Density 0.001%

    No Known Activations