© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8842
    Prev
    Next
    INDEX
    Explanations

    name __

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

    This neuron detects alphanumeric code fragments (strings of letters and digits) that look like transaction or identifier tokens.

    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
     !==
    1.07
    !==
    0.79
     ≠
    0.78
     equal
    0.77
     equals
    0.77
     Equals
    0.77
     Equal
    0.76
     NULL
    0.76
     >
    0.75
    小于
    0.74
    POSITIVE LOGITS
    Howard
    0.55
    Reed
    0.54
    ==(
    0.54
     phối
    0.53
    侩
    0.52
    Ruling
    0.51
    Henderson
    0.51
    timetable
    0.50
    possibly
    0.50
     ኤ
    0.50
    Activations Density 0.002%

    No Known Activations