© 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. 86043
    Prev
    Next
    INDEX
    Explanations

    go followed by a preposition

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

    The neuron strongly activates on the verb “go” (and to a lesser extent on “with”), effectively detecting occurrences of “go” in the text.

    oai_token-act-pair · o4-miniTriggered 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
    {
    -1.73
    买了
    -1.67
     entschieden
    -1.59
     beeinflussen
    -1.59
     revive
    -1.58
    にほんブログ村
    -1.56
     restaurar
    -1.55
     飯店
    -1.55
     एवं
    -1.55
     helicópter
    -1.52
    POSITIVE LOGITS
    ’
    2.31
    ↵
    1.88
    ”
    1.80
     were
    1.73
     They
    1.67
     be
    1.66
    ㉕
    1.63
    ни
    1.61
    “
    1.59
    ⟞
    1.51
    Activations Density 0.005%

    No Known Activations