© 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. 3436
    Prev
    Next
    INDEX
    Explanations

    numbers followed by punctuation or more numbers

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

    The neuron activates on numeric tokens and measurements (e.g., standalone digits or decimal numbers).

    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
    ettek
    0.55
    Abi
    0.55
    Inbox
    0.53
    ӡ
    0.52
    ندگان
    0.51
    ForgotPassword
    0.51
    Catagory
    0.51
    WhiteSpace
    0.50
     Motivational
    0.50
    飡
    0.49
    POSITIVE LOGITS
     lesions
    0.60
     associating
    0.57
     aforesaid
    0.55
     meandering
    0.55
     dieser
    0.53
     above
    0.52
     aforementioned
    0.51
     moiety
    0.50
     itself
    0.49
     derivatives
    0.49
    Activations Density 0.023%

    No Known Activations