© 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. 128649
    Prev
    Next
    INDEX
    Explanations

    The neuron is picking out numeric literals (integer values) in the code.

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

    variable assignment and increment

    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
    の準備
    -1.11
     euró
    -1.07
     rése
    -1.01
     необы
    -1.00
    unjang
    -0.96
    corrhi
    -0.96
     Partager
    -0.96
     meleg
    -0.95
     páns
    -0.94
    に入れて
    -0.92
    POSITIVE LOGITS
     ''
    0.89
     '',
    0.85
    后者
    0.85
    atouille
    0.84
    两者
    0.84
     =
    0.82
     nunmehr
    0.82
    与此同时
    0.82
     puer
    0.80
     PHD
    0.79
    Activations Density 0.001%

    No Known Activations