© 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. 110949
    Prev
    Next
    INDEX
    Explanations

    critical leading to important terms

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

    The neuron primarily detects occurrences of the word “critical.”

    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
    餮
    -0.97
     Should
    -0.96
     weiteren
    -0.93
    侮
    -0.88
     cualquiera
    -0.87
    anyak
    -0.84
    辦法
    -0.82
    PREMI
    -0.81
    мор
    -0.81
     They
    -0.80
    POSITIVE LOGITS
     critical
    1.88
    critical
    1.67
    Critical
    1.62
     Critical
    1.55
     CRITICAL
    1.52
     juncture
    1.44
    ity
    1.30
    CRITICAL
    1.26
     thinking
    1.21
     critically
    1.14
    Activations Density 0.019%

    No Known Activations