© 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. 22-GEMMASCOPE-RES-131K
    7. 25135
    Prev
    Next
    INDEX
    Explanations

    actions and commands

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

    The neuron fires on salient content words—especially technical terms, identifiers, and other concept-bearing tokens (like API names, commands, and key nouns/verbs)—while ignoring common function words and punctuation.

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    istemas
    -0.98
    clairage
    -0.91
    Introdução
    -0.85
     architectures
    -0.83
    ômico
    -0.83
    vrigt
    -0.81
    ujan
    -0.80
    Modelos
    -0.80
    databind
    -0.79
    \}^{
    -0.79
    POSITIVE LOGITS
     actions
    1.19
     commands
    1.09
     action
    1.06
    action
    0.96
    enum
    0.91
     such
    0.91
    actions
    0.90
    ственными
    0.88
    主な
    0.88
    enums
    0.88
    Activations Density 0.152%

    No Known Activations