© 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. 2119
    Prev
    Next
    INDEX
    Explanations

    language identifiers

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

    This neuron detects structural or thematic signpost words that introduce a primary point—e.g. "main," "focus," "cause," "difference," "basis," "goal," or "theme."

    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
    לא
    0.66
    only
    0.66
    Chick
    0.66
    better
    0.65
    ].
    0.64
    Only
    0.63
    Won
    0.63
    گونه
    0.62
    ئے۔
    0.61
    ゴ
    0.61
    POSITIVE LOGITS
     is
    1.22
     είναι
    1.12
     adalah
    0.97
    はこの
    0.97
     هذا
    0.95
     this
    0.94
     dieses
    0.92
     acest
    0.92
     was
    0.91
     dieser
    0.90
    Activations Density 0.340%

    No Known Activations