© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10405
    Prev
    Next
    INDEX
    Explanations

    printable and programmable

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

    the word "printable" or "programmable" in technical or product documentation contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on domain‐specific identifier tokens—i.e. technical labels, code or API names, and configuration property names.

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

    adjectives ending in -able/-ible that denote capabilities, especially in technical or programming contexts.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ↵
    -3.14
     the
    -2.58
     I
    -2.50
     intrepid
    -2.47
    ఽ
    -2.42
     avec
    -2.38
    am
    -2.38
     i
    -2.36
     dwie
    -2.34
     `
    -2.31
    POSITIVE LOGITS
     esmeralda
    3.02
    葙
    2.81
    Getting
    2.80
    what
    2.58
     malheure
    2.56
    櫚
    2.56
    Using
    2.50
    Make
    2.45
    Having
    2.39
     asiatique
    2.39
    Activations Density 0.002%

    No Known Activations