© 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. 34-GEMMASCOPE-RES-131K
    7. 1211
    Prev
    Next
    INDEX
    Explanations

    date and time formats

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

    This neuron fires strongly on the “{” token (and similar block-opening punctuation) that begins code or data structure literals.

    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
     before
    -0.99
     in
    -0.95
     despite
    -0.90
     does
    -0.88
     as
    -0.84
     from
    -0.83
    Before
    -0.82
     after
    -0.81
    ginate
    -0.80
     every
    -0.79
    POSITIVE LOGITS
    voi
    0.97
     XG
    0.96
     GETS
    0.93
     рассказал
    0.92
     ornate
    0.91
    Methode
    0.91
     considerare
    0.90
     Matilda
    0.90
     Helps
    0.89
     NEEDED
    0.88
    Activations Density 0.000%

    No Known Activations