© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 109875
    Prev
    Next
    INDEX
    Explanations

    carve, cut, engrave

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     printed
    -1.07
     PRINTED
    -0.81
     mosa
    -0.78
     constructed
    -0.78
     Werth
    -0.76
     pasteur
    -0.76
    随即
    -0.75
    printed
    -0.74
     trapped
    -0.73
    껑
    -0.71
    POSITIVE LOGITS
     into
    1.62
     carving
    1.19
     carve
    1.16
     deeper
    1.13
     deep
    1.13
     indelible
    1.09
     in
    1.04
     sâu
    0.99
     onto
    0.98
     initials
    0.98
    Activations Density 0.026%

    No Known Activations