© 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. 34-GEMMASCOPE-RES-131K
    4. 66035
    Prev
    Next
    INDEX
    Explanations

    removal or reduction

    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_34/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
     lingue
    -0.80
    cised
    -0.75
    <0xC8>
    -0.74
     stability
    -0.73
    mayo
    -0.72
     stabilizing
    -0.71
    的气息
    -0.69
    わりに
    -0.68
    Restaurante
    -0.68
    さて
    -0.67
    POSITIVE LOGITS
     clutter
    1.16
    burden
    0.94
     stress
    0.89
     fog
    0.89
     fuzz
    0.88
     frizz
    0.87
     burden
    0.83
     fluff
    0.82
     de
    0.81
     congestion
    0.81
    Activations Density 0.053%

    No Known Activations