© 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. 100724
    Prev
    Next
    INDEX
    Explanations

    further X or without ado

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

    The neuron reliably activates on the word “further.”

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

    The neuron is highly sensitive to the adverb “further” (especially in the common phrase “without further ado”).

    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
    WHY
    -1.69
    doesn
    -1.59
     REALLY
    -1.59
     ABSOL
    -1.56
     Tô
    -1.52
    cuz
    -1.50
     VERY
    -1.49
    partly
    -1.49
    pity
    -1.45
    AutoModerator
    -1.44
    POSITIVE LOGITS
     bądź
    1.53
     from
    1.44
    AutoresizingMask
    1.41
    h
    1.39
    ↵
    1.34
    5
    1.32
     would
    1.32
     there
    1.31
     purchase
    1.31
     decided
    1.30
    Activations Density 0.015%

    No Known Activations