© 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. 48776
    Prev
    Next
    INDEX
    Explanations

    activity after an event

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

    The neuron is highly activated by rare or otherwise unusual tokens—i.e. words and names that occur infrequently in the corpus.

    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
    喻
    -0.74
    ‛
    -0.70
     zasi
    -0.69
     largest
    -0.68
     convidados
    -0.68
    циональный
    -0.67
    #@
    -0.66
    masalahan
    -0.66
    Longitude
    -0.66
    Continuous
    -0.66
    POSITIVE LOGITS
    annya
    0.94
    Brien
    0.81
    看去
    0.78
     convencer
    0.76
     ANC
    0.74
     BZ
    0.73
     adhe
    0.72
     studen
    0.72
    钛
    0.71
    IntPtr
    0.71
    Activations Density 0.023%

    No Known Activations