© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129678
    Prev
    Next
    INDEX
    Explanations

    numbers and foreign words

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

    The neuron fires on subjective, evaluative words—strong descriptive terms that convey quality or change (e.g. innovation, improved, fierce, diverse, announcements, populism).

    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
     then
    -3.00
     Then
    -1.70
     they
    -1.59
    </i>
    -1.52
     who
    -1.49
    Then
    -1.49
     kemudian
    -1.47
     затем
    -1.46
    then
    -1.46
    ❣️
    -1.39
    POSITIVE LOGITS
     zweier
    1.49
     viena
    1.45
     recientemente
    1.43
     pirm
    1.40
    Ϟ
    1.39
     कई
    1.38
     jednym
    1.37
     ثلاثة
    1.36
    geoning
    1.35
    .”.
    1.35
    Activations Density 0.088%

    No Known Activations