© 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. 24-GEMMASCOPE-2-RES-16K
    7. 8371
    Prev
    Next
    INDEX
    Explanations

    asking to know

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

    The neuron is keyed to sentence-initial positive/hopeful discourse markers (e.g. “Fortunately,” “Luckily,” “Glad,” “Perfect,” etc.) that signal a shift or positive evaluation.

    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_24_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
     WOULD
    0.77
    どうしても
    0.77
     certainly
    0.75
     will
    0.71
     would
    0.71
     inappropriately
    0.69
     WILL
    0.68
     sẽ
    0.67
    куру
    0.66
     попро
    0.64
    POSITIVE LOGITS
     удалось
    0.81
     sodass
    0.81
     udało
    0.80
    👍
    0.80
     thankfully
    0.79
     उन्होंने
    0.78
     zodat
    0.77
     siker
    0.77
     덕
    0.76
    🙏🏻
    0.76
    Activations Density 0.409%

    No Known Activations