© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6904
    Prev
    Next
    INDEX
    Explanations

    general techniques, for the, in interplay

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

    This neuron mainly detects numeric expressions and their associated unit or multiplier words (e.g. “3 times”).

    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
    з
    0.81
    ﺑ
    0.80
    ಟೆ
    0.73
    ב
    0.73
    டன்
    0.71
    가가
    0.71
    ﻓ
    0.71
    ble
    0.69
     кноп
    0.69
    zinha
    0.68
    POSITIVE LOGITS
     procrast
    0.85
    始め
    0.80
    দৃষ্টি
    0.80
     covariant
    0.80
     nutshell
    0.79
    ()=>{
    0.78
     loudness
    0.76
     covariates
    0.75
     uncooked
    0.75
     centrifug
    0.74
    Activations Density 0.000%

    No Known Activations