© 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. 2573
    Prev
    Next
    INDEX
    Explanations

    seriously, need, or, worse

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

    The neuron fires on strongly positive evaluative or praising language (e.g. “best,” “trust,” “very hard to find”).

    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
    0.68
    -
    0.62
     reported
    0.49
    Continue
    0.49
     memperoleh
    0.48
    ‐
    0.47
     düzey
    0.47
    '
    0.46
    ‑
    0.46
     valamint
    0.46
    POSITIVE LOGITS
    exams
    0.77
    ecção
    0.70
     Quadrupèdes
    0.70
    sächlich
    0.69
    abilă
    0.68
    allenges
    0.67
    ucking
    0.67
    adrat
    0.67
    esquerda
    0.66
    éseket
    0.66
    Activations Density 0.449%

    No Known Activations