© 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. 6678
    Prev
    Next
    INDEX
    Explanations

    enroll multiple fingerprints (at least 5-7 is a good target)

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

    The neuron activates on floating‐point numeric timestamps (those decimal numbers that appear in date/time or metadata fields).

    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
    Y
    1.02
    Z
    1.00
    F
    0.94
    E
    0.92
    Od
    0.89
    EN
    0.85
    ON
    0.83
    de
    0.81
    Ot
    0.81
    On
    0.80
    POSITIVE LOGITS
     стоят
    0.90
     hangout
    0.88
     зимой
    0.86
     esperando
    0.84
     обраща
    0.83
     leve
    0.81
     কিনে
    0.81
     этого
    0.79
     oversee
    0.79
     hustle
    0.78
    Activations Density 0.000%

    No Known Activations