© 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. 64742
    Prev
    Next
    INDEX
    Explanations

    dataframe objects

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

    The neuron fires on numeric literals (various digit tokens) in code—i.e. it detects numbers.

    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
     tomaron
    -1.04
     cobr
    -0.84
     ye
    -0.81
     again
    -0.80
     at
    -0.79
     from
    -0.79
     early
    -0.79
    Allister
    -0.79
     반
    -0.78
    Commenting
    -0.78
    POSITIVE LOGITS
    ⁉
    1.02
    🙊
    0.94
    }',
    0.93
     Chancery
    0.92
    ométrie
    0.91
    signable
    0.91
    🤏
    0.89
     betere
    0.89
    🍴
    0.88
     regimens
    0.87
    Activations Density 0.018%

    No Known Activations