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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 12165
    Prev
    Next
    INDEX
    Explanations

    method or class names

    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
     alfabe
    -0.90
    suche
    -0.84
    Satur
    -0.84
     hob
    -0.83
    TAWA
    -0.82
    
    -0.82
    EREN
    -0.82
    Itu
    -0.81
    Arbe
    -0.81
     recherches
    -0.80
    POSITIVE LOGITS
     that
    1.23
     our
    1.08
     this
    1.02
     which
    0.97
     difficult
    0.91
    دارة
    0.86
     plaintiffs
    0.86
     dangerous
    0.85
    可能會
    0.85
    してない
    0.85
    Activations Density 0.009%

    No Known Activations