© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 115600
    Prev
    Next
    INDEX
    Explanations

    unsafe keywords

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

    The neuron preferentially activates on longer, content-bearing words (e.g. multi-syllabic verbs and nouns) rather than short function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     jacuzzi
    -1.03
     Jacuzzi
    -0.89
     Hot
    -0.80
    )});
    -0.78
    还是很
    -0.75
    enne
    -0.74
    ミラ
    -0.74
    voll
    -0.72
    小白
    -0.72
    
    -0.71
    POSITIVE LOGITS
    𝑤
    0.80
    tus
    0.76
     monologue
    0.75
    

    0.73
     sejumlah
    0.73
     продук
    0.72
    ∭
    0.70
     fortnight
    0.70
    новения
    0.69
     but
    0.69
    Activations Density 0.004%

    No Known Activations