© 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. 86642
    Prev
    Next
    INDEX
    Explanations

    Bundes compounds and busy periods

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

    The neuron primarily detects longer, multi-syllable content words (e.g. snowflake, fermentation, courthouse, whirlwind, Bundes).

    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
    7
    -3.42
    6
    -3.11
    2
    -3.08
    8
    -3.06
    4
    -3.05
    ing
    -2.98
    9
    -2.78
    3
    -2.78
    5
    -2.77
    You
    -2.72
    POSITIVE LOGITS
     concorda
    3.08
     nasceu
    3.03
    饹
    2.89
    圌
    2.77
    螃
    2.77
     namorados
    2.66
    fuels
    2.59
    鉏
    2.58
    蚍
    2.58
    僰
    2.52
    Activations Density 0.038%

    No Known Activations