© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 21740
    Prev
    Next
    INDEX
    Explanations

    must go hand in hand

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

    This neuron responds to contrastive or emphatic discourse markers—short adverbs and connectors like “but,” “rather,” “exactly,” etc., that signal emphasis or a shift in tone.

    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
    ,
    -1.36
    ícias
    -1.32
    他
    -1.27
     each
    -1.25
    <td>
    -1.25
    ruptcy
    -1.23
     those
    -1.16
    lized
    -1.12
    .)-
    -1.12
     LIABLE
    -1.11
    POSITIVE LOGITS
     което
    1.37
     различни
    1.36
     podczas
    1.33
    muñ
    1.25
     cuarent
    1.23
     algunas
    1.23
     mépris
    1.23
    杁
    1.22
     بعضی
    1.22
    となっています
    1.20
    Activations Density 0.070%

    No Known Activations