© 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. 12469
    Prev
    Next
    INDEX
    Explanations

    numbers and years

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

    The neuron peaks on the very first content word of each new text segment, effectively marking the start of a passage.

    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
    Crawler
    0.63
    Sur
    0.60
    Trajectory
    0.57
    ὶ
    0.55
    爬
    0.55
    Strings
    0.54
    Fault
    0.54
    任何
    0.53
     到
    0.53
     Maturity
    0.53
    POSITIVE LOGITS
    ullen
    0.59
    wet
    0.55
    க்காக
    0.55
     foreach
    0.54
    summon
    0.54
     fios
    0.53
    isSelected
    0.53
     ላይ
    0.52
    নেই
    0.52
    ంతరం
    0.52
    Activations Density 0.000%

    No Known Activations