© 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. 72351
    Prev
    Next
    INDEX
    Explanations

    when called

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

    The neuron fires on the word “called,” i.e. it detects occurrences of the token “called.”

    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
     pomocí
    -1.07
     his
    -1.07
    starting
    -1.05
    礎
    -1.02
     Duits
    -1.00
     its
    -0.96
    衬衫
    -0.95
     ortam
    -0.94
     nasz
    -0.94
    suited
    -0.93
    POSITIVE LOGITS
     when
    2.67
     after
    1.91
     by
    1.54
     before
    1.47
     during
    1.43
     every
    1.41
     if
    1.38
     cuando
    1.38
    when
    1.35
     whenever
    1.34
    Activations Density 0.021%

    No Known Activations