© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5158
    Prev
    Next
    INDEX
    Explanations

    technical terms or conditional expressions

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

    The neuron detects imperative or instructional phrasing—that is, verbs and words used to give directions or commands.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    insurer
    0.40
    ָד
    0.39
    iritual
    0.38
    農
    0.38
    ジタル
    0.37
    微生物
    0.36
    聖
    0.36
    IVERSITY
    0.36
    土
    0.36
    ocultural
    0.35
    POSITIVE LOGITS
     elenco
    0.44
     τη
    0.39
     gama
    0.37
     acho
    0.37
     peuple
    0.37
     tinha
    0.36
    idän
    0.36
    随即
    0.36
     melee
    0.36
     tane
    0.36
    Activations Density 1.587%

    No Known Activations