© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104054
    Prev
    Next
    INDEX
    Explanations

    greetings and questions

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

    The neuron activates on speaker‐attribution markers—i.e. names or identifiers immediately followed by a colon, indicating a change of speaker or a transcript cue.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     a
    -2.19
     "
    -2.02
    一旁的
    -1.70
     an
    -1.61
    ти
    -1.60
     of
    -1.59
     new
    -1.55
     on
    -1.54
     many
    -1.53
     both
    -1.46
    POSITIVE LOGITS
     Wikime
    1.88
     Unterscheidung
    1.63
    ilers
    1.60
     presentada
    1.55
     affez
    1.55
     Darstellung
    1.55
     Hinsicht
    1.51
    ーん
    1.50
    
    1.47
    ровке
    1.46
    Activations Density 0.003%

    No Known Activations