© 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. 22-GEMMASCOPE-RES-131K
    7. 34673
    Prev
    Next
    INDEX
    Explanations

    French and German words

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

    The neuron specializes in detecting significant leading whitespace or indentation levels (i.e. blocks of spaces/tabs at the start of lines).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     will
    -1.33
     to
    -1.16
     can
    -1.14
     would
    -1.09
     required
    -1.05
     before
    -1.05
     should
    -0.98
     turn
    -0.96
     uses
    -0.96
     At
    -0.94
    POSITIVE LOGITS
     équilibre
    1.14
     époux
    1.10
     véhicule
    1.05
     refroid
    1.05
    Ona
    1.02
     trhu
    1.00
    ➀
    0.99
    weiß
    0.98
    recen
    0.98
     confé
    0.98
    Activations Density 0.005%

    No Known Activations