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. 22-GEMMASCOPE-RES-131K
    4. 11977
    Prev
    Next
    INDEX
    Explanations

    as a descriptor

    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
     existing
    -0.84
    drawiam
    -0.82
     maillot
    -0.79
     Secretaries
    -0.79
    Attacking
    -0.79
     ambiental
    -0.78
     tournant
    -0.78
    Extraction
    -0.77
    Inves
    -0.76
    Twe
    -0.76
    POSITIVE LOGITS
     from
    0.91
    名
    0.87
     也
    0.84
     Join
    0.82
     специаль
    0.82
    bootstrap
    0.82
    HANDLE
    0.82
     in
    0.80
    Join
    0.80
    rowIndex
    0.79
    Activations Density 0.060%

    No Known Activations