© 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. 11465
    Prev
    Next
    INDEX
    Explanations

    Customers who bought this also bought

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

    The neuron is detecting numerical metrics (e.g. view counts or other numeric values) in the text.

    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
    Aaj
    0.73
    手指
    0.71
    Diagnostics
    0.71
    障
    0.69
    <h3>
    0.68
    Eye
    0.67
    Her
    0.64
    Video
    0.63
    杈
    0.63
    ચી
    0.62
    POSITIVE LOGITS
     predomin
    0.80
     want
    0.74
     masters
    0.65
     penthouse
    0.65
     primarily
    0.65
     mert
    0.65
     Myrtle
    0.64
     unwilling
    0.63
     descriptions
    0.63
     salesmen
    0.62
    Activations Density 0.000%

    No Known Activations