© 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. 6329
    Prev
    Next
    INDEX
    Explanations

    imagining or hypothetical situations

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

    This neuron is a “rarity” detector: it spikes on unusually uncommon or low‐frequency tokens (i.e. words you don’t see very often).

    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
    ベージュ
    0.94
    何も
    0.90
     localidades
    0.88
    iformis
    0.87
     mesmas
    0.87
    ﺤ
    0.84
    BINARY
    0.81
    従業
    0.81
    bildung
    0.81
     রোগী
    0.80
    POSITIVE LOGITS
    ,
    0.75
     CL
    0.71
     Shop
    0.69
     Che
    0.66
     '(
    0.66
     (
    0.66
     Cell
    0.66
     Pins
    0.65
     раз
    0.65
     jt
    0.64
    Activations Density 0.000%

    No Known Activations