© 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. 9119
    Prev
    Next
    INDEX
    Explanations

    A1

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

    The neuron activates on spans of digits (often mixed with letters or punctuation) that look like section/question numbers or other alphanumeric identifiers/codes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    alphanumeric product codes, model numbers, and technical identifiers.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    alphanumeric identifiers and codes (e.g., model numbers, SKUs, barcodes, and sectioned Q/A numbering with digits and periods).

    oai_token-act-pair · gpt-5Triggered 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
     परमा
    0.63
     अन्याय
    0.63
     repress
    0.58
     للق
    0.58
    ຯ
    0.57
     яе
    0.56
    estries
    0.55
     bygone
    0.55
    යින්
    0.55
     atribu
    0.54
    POSITIVE LOGITS
    R
    0.87
    E
    0.77
    Agg
    0.73
    S
    0.72
    A
    0.71
    G
    0.71
    I
    0.69
    Hed
    0.69
    M
    0.68
    AG
    0.67
    Activations Density 0.221%

    No Known Activations