© 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. 129267
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on literal zero tokens (the numeral “0”) in code, especially when used in bit‐masking expressions.

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

    special entities and phrases

    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
    などを
    -1.34
    などが
    -1.31
    licitations
    -1.30
     münchen
    -1.16
    が上が
    -1.13
    などの
    -1.13
    があり
    -1.11
    を
    -1.10
    が
    -1.09
    vereign
    -1.08
    POSITIVE LOGITS
    に戻る
    1.19
    感じで
    1.16
    ところは
    1.07
    為に
    1.01
     manchmal
    1.00
    のでは
    0.96
    ところです
    0.95
    ってる
    0.95
    🦧
    0.94
    izzano
    0.93
    Activations Density 0.001%

    No Known Activations