© 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. 92821
    Prev
    Next
    INDEX
    Explanations

    connecting actions and communication

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

    The neuron strongly responds to very common English words—especially frequent function words (e.g. “and,” “do,” “about,” “or,” “are”) and other high-frequency 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
     Verbal
    -0.98
     Vocals
    -0.92
    ละเอียด
    -0.82
    regado
    -0.82
     società
    -0.82
     Vocal
    -0.81
     Mortality
    -0.79
    sizable
    -0.78
    言葉を
    -0.78
     vēl
    -0.77
    POSITIVE LOGITS
     actions
    1.44
     gesture
    1.39
     gestures
    1.29
     action
    1.09
     writing
    1.08
     gestos
    1.05
     written
    1.02
    gesture
    1.01
     pictures
    0.97
     deeds
    0.95
    Activations Density 0.045%

    No Known Activations