© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 79543
    Prev
    Next
    INDEX
    Explanations

    quotation marks and commas

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

    The neuron activates on colon characters that introduce or follow speech‐attribution verbs (i.e. the “said:” or “explained:” before a quotation).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    リム
    -0.77
    分支
    -0.76
    ysteme
    -0.76
    '],
    -0.75
    螞
    -0.75
    /',
    -0.75
    šiu
    -0.74
     menyebutkan
    -0.73
    více
    -0.73
    wut
    -0.71
    POSITIVE LOGITS
     செ
    0.93
     """
    0.84
     I
    0.82
    .“
    0.77
     said
    0.77
    شرة
    0.75
     "
    0.75
     eingeführt
    0.74
    colata
    0.74
     stück
    0.74
    Activations Density 0.032%

    No Known Activations