© 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. 6866
    Prev
    Next
    INDEX
    Explanations

    URL HTTPURL

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

    This neuron responds to tokens containing numbers or alphanumeric codes (e.g. barcodes, IDs, statistics).

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

    alphanumeric codes and identifiers, particularly those with mixed letters and numbers.

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

    acronyms and other all-caps or alphanumeric abbreviations/codes.

    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
     taas
    0.56
    试试
    0.55
    isol
    0.55
     innumerable
    0.55
     ritor
    0.55
     legy
    0.55
    todos
    0.54
     kroner
    0.54
     détente
    0.54
     moat
    0.53
    POSITIVE LOGITS
     Quite
    0.67
     Nearly
    0.65
     Pretty
    0.63
    ITTLE
    0.62
     Like
    0.60
    ENING
    0.58
     Yearly
    0.57
     Quick
    0.56
     Seab
    0.56
    )\$(
    0.56
    Activations Density 0.379%

    No Known Activations