© 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. 123473
    Prev
    Next
    INDEX
    Explanations

    call for

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

    This neuron fires strongly on the very first content word in each new text segment.

    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
    Either
    -1.56
    Their
    -1.51
    either
    -1.48
     Estamos
    -1.44
    либо
    -1.41
     "'");
    -1.41
     either
    -1.39
     Tenemos
    -1.32
     Их
    -1.32
     Either
    -1.31
    POSITIVE LOGITS
    0
    1.41
    3
    1.41
     like
    1.40
    versy
    1.37
    quelize
    1.34
     would
    1.33
    1
    1.20
     (
    1.18
    天子
    1.14
     by
    1.12
    Activations Density 0.205%

    No Known Activations