© 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. 10-GEMMASCOPE-RES-131K
    7. 111579
    Prev
    Next
    INDEX
    Explanations

    do/did/does verb constructions

    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
     just
    -1.76
     to
    -1.75
     not
    -1.73
     never
    -1.60
     also
    -1.37
     either
    -1.36
    4
    -1.34
    .,
    -1.32
    なども
    -1.30
     but
    -1.28
    POSITIVE LOGITS
    Karakteristik
    1.69
     habia
    1.49
    Kelebihan
    1.48
     Autó
    1.41
     aprendido
    1.41
     adhesivo
    1.41
    牁
    1.40
    ᚺ
    1.39
    넋
    1.39
    鰱
    1.38
    Activations Density 0.035%

    No Known Activations