© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9494
    Prev
    Next
    INDEX
    Explanations

    `(` or `$` followed by number or `(`

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

    The neuron flags unusually long or low‐frequency words—i.e. rare vocabulary items.

    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_31_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
     ইহাতে
    0.74
     黃
    0.70
    preview
    0.70
     GOLD
    0.69
     ===============
    0.69
     equals
    0.68
     Ну
    0.67
     Equal
    0.66
     Igual
    0.66
     Calculates
    0.66
    POSITIVE LOGITS
    噎
    0.77
    ]，
    0.72
     domaines
    0.72
    ugat
    0.71
    ต่ำ
    0.70
    美元
    0.68
     situado
    0.68
    সীম
    0.68
    臓
    0.67
    糗
    0.67
    Activations Density 0.000%

    No Known Activations