© 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. 34-GEMMASCOPE-RES-131K
    7. 108873
    Prev
    Next
    INDEX
    Explanations

    docker build and run

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

    This neuron fires on runs of punctuation (e.g. “..” directives or “-”/“--” flags), i.e. tokens composed entirely of repeated punctuation marks.

    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
    DAY
    -0.81
    onChangeText
    -0.76
    mbangkan
    -0.72
    andes
    -0.71
    née
    -0.71
    TRA
    -0.69
     Dayton
    -0.69
    љ
    -0.69
    dinger
    -0.69
     dirección
    -0.68
    POSITIVE LOGITS
     Container
    0.71
     allot
    0.65
    ⟡
    0.64
    ocumented
    0.64
     Tox
    0.63
     đấu
    0.63
    批量
    0.63
     apologe
    0.62
    container
    0.61
    zema
    0.61
    Activations Density 0.065%

    No Known Activations