© 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. 79415
    Prev
    Next
    INDEX
    Explanations

    starts of phrases

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

    The neuron strongly activates on whitespace and punctuation tokens (e.g. spaces, periods, newlines) rather than on actual words.

    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
     特
    -1.02
    чника
    -1.00
     offizi
    -0.85
    іль
    -0.84
     būt
    -0.83
     réguli
    -0.82
     cuire
    -0.82
     tangent
    -0.81
     ufficiali
    -0.79
     anzuzeigen
    -0.79
    POSITIVE LOGITS
     yourselves
    1.11
     aliquot
    0.90
     communal
    0.86
     chiefly
    0.85
     Biophys
    0.84
     竜
    0.83
    PropertyChanging
    0.83
    vaig
    0.82
     preço
    0.82
    얼
    0.82
    Activations Density 0.020%

    No Known Activations