© 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. 1097
    Prev
    Next
    INDEX
    Explanations

    immediate, key, understanding, learning, literal, retention

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

    The neuron is primarily detecting numerical and quantitative tokens (e.g. digits, measurements, stats) in the text.

    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
    ҳ
    0.55
    ...?
    0.52
    ...</
    0.51
     확보
    0.51
     rigorously
    0.49
    ӯ
    0.49
    ...!
    0.48
    ...\
    0.48
     ...,
    0.48
     sectoral
    0.48
    POSITIVE LOGITS
     Một
    0.71
    hygine
    0.70
    𝙔
    0.70
     근데
    0.68
    little
    0.66
    رسٹ
    0.65
     Any
    0.65
     Also
    0.64
    ریشن
    0.63
    locking
    0.63
    Activations Density 0.122%

    No Known Activations