Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 222
    Prev
    Next
    INDEX
    Explanations

    introduces statements with "So"

    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
     is
    -2.06
     "
    -1.74
     y
    -1.73
     
    -1.70
     of
    -1.67
    ↵↵
    -1.58
     acercarse
    -1.58
    潆
    -1.53
     You
    -1.53
     และ
    -1.52
    POSITIVE LOGITS
    ⸰
    1.93
     infal
    1.91
    ۰۰
    1.81
     esat
    1.80
     monstru
    1.80
     tremend
    1.73
     concr
    1.73
     precau
    1.66
    ㊗
    1.66
     metic
    1.65
    Activations Density 0.023%

    No Known Activations