© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6506
    Prev
    Next
    INDEX
    Explanations

    feature comparisons

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

    This neuron fires on cautionary or strongly negative customer‐service statements—warnings, directives, or complaints about staff behavior (e.g. do not pay, staff deceit, terrible service).

    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_31_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
     Everyone
    0.65
     *"
    0.63
     [_
    0.63
     situazioni
    0.63
    椃
    0.63
    ⁢
    0.62
     *\
    0.60
     beğen
    0.60
     ඉතා
    0.59
    POSITIVE LOGITS
     travelled
    0.57
     undertook
    0.55
    Spent
    0.55
     spent
    0.53
     wybra
    0.52
     volunteered
    0.50
     каких
    0.50
     leapt
    0.50
    选择了
    0.49
     heur
    0.49
    Activations Density 0.046%

    No Known Activations