© 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. 22-GEMMASCOPE-RES-131K
    7. 26086
    Prev
    Next
    INDEX
    Explanations

    timed decisions or outcomes

    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
     vários
    -1.45
     célè
    -1.30
    指着
    -1.30
    depuis
    -1.27
    actéristi
    -1.27
     besonder
    -1.25
     várias
    -1.25
     probablement
    -1.24
     ་
    -1.23
    キラキラ
    -1.23
    POSITIVE LOGITS
     first
    2.14
     with
    1.79
     when
    1.74
     by
    1.67
     two
    1.50
     came
    1.48
     shortly
    1.48
     Shortly
    1.45
     out
    1.41
     When
    1.39
    Activations Density 0.006%

    No Known Activations