© 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. 11907
    Prev
    Next
    INDEX
    Explanations

    file and directory structure

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

    The neuron primarily activates on lines marked as additions in diff/patch output (i.e. lines beginning with “+”).

    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
    minus
    0.87
    Minus
    0.86
     minus
    0.85
     Minus
    0.83
     минус
    0.77
    ()-
    0.66
     deducted
    0.64
    蓝牙
    0.64
     −
    0.64
    Subtract
    0.64
    POSITIVE LOGITS
     invites
    0.50
     cracker
    0.50
     invoc
    0.50
     opinião
    0.50
     can
    0.50
     crackers
    0.50
    водить
    0.50
     was
    0.48
     mentale
    0.47
     evidence
    0.47
    Activations Density 0.002%

    No Known Activations