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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 40255
    Prev
    Next
    INDEX
    Explanations

    limitations in data

    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_22/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
    ⁉
    -1.41
    🤧
    -1.41
     marts
    -1.37
    ❣
    -1.37
     although
    -1.37
    🤨
    -1.36
     lune
    -1.32
     tribus
    -1.32
     dets
    -1.30
     sista
    -1.30
    POSITIVE LOGITS
     my
    1.90
     you
    1.75
     możemy
    1.73
     I
    1.72
     our
    1.61
     tutaj
    1.58
     fantastic
    1.55
     znajdu
    1.52
     incredibly
    1.48
    可能
    1.47
    Activations Density 0.009%

    No Known Activations