© 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. 58133
    Prev
    Next
    INDEX
    Explanations

    canned food items

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

    The neuron fires on numeric tokens—quantities, measurements, counts, and other numbers in the text.

    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
    tott
    -0.93
    haan
    -0.79
    lnie
    -0.76
    liefer
    -0.74
    SNMP
    -0.73
    roek
    -0.73
     Burnett
    -0.73
    尔夫
    -0.73
    èvres
    -0.72
     Corrections
    -0.72
    POSITIVE LOGITS
     canned
    2.47
     cans
    2.30
     Canned
    1.84
    缶
    1.52
     Cans
    1.52
    罐
    1.47
    cans
    1.44
     tin
    1.39
     canning
    1.37
     консер
    1.29
    Activations Density 0.027%

    No Known Activations