© 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. 22-GEMMASCOPE-RES-131K
    7. 64651
    Prev
    Next
    INDEX
    Explanations

    verbs and pronouns initiating phrases

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

    The neuron activates on capitalized multiword phrases that are song or track titles in a music listing.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     jūsų
    -1.47
     naszych
    -1.46
     erhi
    -1.45
     abbigliamento
    -1.40
     Eindruck
    -1.39
     mojej
    -1.36
     exemplaires
    -1.34
    voici
    -1.32
     naszego
    -1.31
     dlaczego
    -1.31
    POSITIVE LOGITS
     There
    1.48
     What
    1.41
     They
    1.41
     where
    1.37
     when
    1.31
     which
    1.31
     Some
    1.28
     If
    1.27
     Often
    1.22
    咆
    1.21
    Activations Density 0.054%

    No Known Activations