© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 40128
    Prev
    Next
    INDEX
    Explanations

    want to do something

    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_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
     further
    -1.39
     solely
    -1.24
     entirely
    -1.21
     盛り
    -1.20
     simply
    -1.20
     only
    -1.19
     greatly
    -1.16
     will
    -1.16
    NSIndexPath
    -1.16
     elaborate
    -1.14
    POSITIVE LOGITS
     tých
    1.49
    最初に
    1.42
    these
    1.42
    those
    1.40
     dobré
    1.34
    and
    1.31
     iemand
    1.30
     piaci
    1.24
     to
    1.24
    küche
    1.23
    Activations Density 0.031%

    No Known Activations