© 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. 34-GEMMASCOPE-RES-131K
    7. 58186
    Prev
    Next
    INDEX
    Explanations

    ignore or disregard

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

    The neuron fires most strongly on literal numeric tokens (i.e. digits or multi‐digit 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_34/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
    dehydes
    -0.73
    Transk
    -0.70
    ufu
    -0.70
    };*/
    -0.69
    sedown
    -0.69
    CastException
    -0.68
    観音
    -0.67
    HostException
    -0.67
    idity
    -0.67
    undrum
    -0.67
    POSITIVE LOGITS
     ignore
    6.88
    ignore
    6.00
     ignored
    6.00
    Ignore
    5.78
     ignores
    5.69
     Ignore
    5.66
     ignoring
    5.63
    忽略
    4.66
    ignored
    4.59
    Ignoring
    4.56
    Activations Density 0.231%

    No Known Activations