© 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. 329
    Prev
    Next
    INDEX
    Explanations

    the body's response

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron identifies the beginning of quoted text or passages in brackets, particularly at the start of news articles or academic content.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    ]].
    -1.09
    ⠕
    -1.00
    kto
    -0.98
    ·
    -0.97
    kalk
    -0.93
     Тогда
    -0.91
     falou
    -0.88
    ffron
    -0.88
    pecially
    -0.88
    versation
    -0.88
    POSITIVE LOGITS
     having
    1.21
     for
    1.08
     getting
    1.08
     using
    1.08
     within
    1.06
     if
    1.05
     because
    1.05
     there
    1.02
     through
    0.98
     lintas
    0.97
    Activations Density 0.000%

    No Known Activations