© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25776
    Prev
    Next
    INDEX
    Explanations

    medical, sports, or technical terms and idioms

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

    numerical values, mathematical expressions, and mathematical concepts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is specifically keyed to the token “musician,” activating strongly whenever that word (indicating someone’s role as a musician) appears.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    くなります
    -1.01
    Ꭵ
    -0.96
     nejen
    -0.93
    これらの
    -0.89
    ريبة
    -0.87
     and
    -0.85
    eding
    -0.84
    向了
    -0.84
    aliment
    -0.84
    욘
    -0.84
    POSITIVE LOGITS
    はどう
    0.85
     eeuw
    0.84
    фин
    0.84
     COND
    0.82
     fiatal
    0.82
    ְב
    0.81
     HARRY
    0.80
    Passcode
    0.79
     we
    0.79
    gerea
    0.79
    Activations Density 0.007%

    No Known Activations