© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 64021
    Prev
    Next
    INDEX
    Explanations

    perform a function

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

    The neuron is essentially flagging very common, high‐frequency tokens (both frequent function words and common content words).

    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
    依靠
    -0.79
    Pozri
    -0.79
     fejles
    -0.78
    杼
    -0.78
    Strand
    -0.74
     ミュージ
    -0.73
    ΩΝ
    -0.73
     endeavoured
    -0.72
    ِن
    -0.71
    ternut
    -0.71
    POSITIVE LOGITS
     performing
    4.69
     perform
    4.47
     performs
    4.09
    Performing
    3.88
    performing
    3.81
    Perform
    3.80
    perform
    3.69
     PERFORM
    3.66
     Performing
    3.66
     выпол
    3.56
    Activations Density 0.164%

    No Known Activations