© 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. 75226
    Prev
    Next
    INDEX
    Explanations

    any or too

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

    This neuron detects words and phrases that express prohibition or preventing something (e.g. ensuring none talk, don’t like the public peeking).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     plagio
    -0.98
     ume
    -0.97
     eser
    -0.94
     prejudices
    -0.91
     Martí
    -0.90
    ziej
    -0.89
     totalitarian
    -0.86
     jaro
    -0.85
    arshal
    -0.85
    しゃぶ
    -0.85
    POSITIVE LOGITS
     too
    1.41
     any
    1.36
     слишком
    1.26
    有任何
    1.22
     demasi
    1.02
     anything
    1.00
     groaned
    0.99
     demasiado
    0.92
    任何
    0.90
     qualquer
    0.89
    Activations Density 0.092%

    No Known Activations