© 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. 31-GEMMASCOPE-2-RES-16K
    7. 12707
    Prev
    Next
    INDEX
    Explanations

    ' followed by names or descriptions

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

    The neuron fires on unusually infrequent or domain‐specific tokens (e.g. technical identifiers, proper names, punctuation marks like commas), effectively flagging rare or low‐frequency words.

    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
     मौसम
    0.59
     gastronomy
    0.57
     kecamatan
    0.56
    天氣
    0.53
     currentDate
    0.52
     destinat
    0.52
     municipio
    0.51
     citizenship
    0.50
     เพ
    0.50
     mái
    0.50
    POSITIVE LOGITS
     Description
    0.93
    value
    0.89
     value
    0.89
    对应的
    0.87
     Value
    0.83
    値
    0.83
     description
    0.82
    对应
    0.80
     values
    0.78
     वैल्यू
    0.78
    Activations Density 0.103%

    No Known Activations