© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2261
    Prev
    Next
    INDEX
    Explanations

    The neuron is highly active on tokens that contain digits or non-letter punctuation (e.g. numeric time stamps or words with trailing hyphens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    al
    0.80
    ේෂ
    0.74
    c
    0.74
    *",
    0.70
    هداف
    0.70
    p
    0.69
    TargetPath
    0.68
    b
    0.68
    g
    0.68
    ­
    0.67
    POSITIVE LOGITS
    ativos
    0.81
    㤈
    0.80
     সাহেব
    0.80
    dimensioni
    0.80
    俗
    0.79
     лигасы
    0.77
     peculi
    0.76
    
    0.76
     habitually
    0.76
     stifle
    0.75
    Activations Density 0.000%

    No Known Activations