© 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. 108918
    Prev
    Next
    INDEX
    Explanations

    statements of being

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

    The neuron spikes on very common, high‐frequency tokens (e.g. “it,” “really,” “what,” “I,” “there,” etc.), effectively marking the most frequent words in the text.

    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
     even
    -0.98
     Doing
    -0.92
    NSInteger
    -0.88
     actually
    -0.86
     drzwi
    -0.86
    Even
    -0.83
    Doing
    -0.81
    doing
    -0.79
     didel
    -0.78
     indicates
    -0.77
    POSITIVE LOGITS
     IS
    1.28
     wel
    1.16
     כן
    1.07
     is
    1.05
    IS
    0.97
     есть
    0.96
    venidos
    0.94
     ARE
    0.88
    确实是
    0.86
    is
    0.86
    Activations Density 0.020%

    No Known Activations