© 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. 106194
    Prev
    Next
    INDEX
    Explanations

    consequences or lack of effect

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

    The neuron is essentially detecting very high‐frequency (common) tokens—i.e. function words and other extremely frequent terms.

    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
    eraard
    -0.75
     kazan
    -0.73
     Technik
    -0.73
     knitted
    -0.72
    갑
    -0.71
    лакти
    -0.71
     onError
    -0.70
     Audi
    -0.69
     will
    -0.69
     Yo
    -0.69
    POSITIVE LOGITS
     puzzle
    0.98
     channels
    0.91
    が悪い
    0.84
    sizable
    0.83
    bang
    0.82
    Documentos
    0.81
    movements
    0.81
     Puzzle
    0.80
     concave
    0.80
    ߥ
    0.80
    Activations Density 0.022%

    No Known Activations