© 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. 34-GEMMASCOPE-RES-131K
    7. 64309
    Prev
    Next
    INDEX
    Explanations

    by unit

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

    The neuron strongly fires on the preposition “by” when it appears in repetitive or hyphenated constructs (e.g. “line by line,” “play-by-play,” “one by one,” “round by round”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Can
    -0.78
    แน่น
    -0.73
    •
    -0.69
    斯科
    -0.68
     ©
    -0.68
     хвати
    -0.66
     /
    -0.66
    be
    -0.66
     ‎
    -0.66
     Her
    -0.66
    POSITIVE LOGITS
    々
    1.18
    ๆ
    1.17
     البعض
    1.15
    ៗ
    0.95
     πρ
    0.94
    amation
    0.90
    nutrition
    0.89
    trom
    0.87
     captiv
    0.86
     lauded
    0.86
    Activations Density 0.269%

    No Known Activations