© 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. 58863
    Prev
    Next
    INDEX
    Explanations

    Seeing such useful drugs

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

    phrases indicating visual perception, particularly with words like "seeing" and "saw" when someone notices or observes something.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is strongly triggered by content‐bearing words that describe perception or evaluation (e.g. “Seeing,” “saw,” “confused,” “useful”), i.e. it flags words that carry narrative or descriptive weight.

    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
     to
    -1.08
     we
    -0.94
     they
    -0.92
    ycznego
    -0.87
     Verbrechen
    -0.83
     legislature
    -0.81
     betreffenden
    -0.81
    orari
    -0.79
    gherita
    -0.79
     thanks
    -0.78
    POSITIVE LOGITS
    如此
    1.05
     stiller
    0.96
     таком
    0.96
     clearly
    0.95
     behaving
    0.94
    clearly
    0.93
     Safran
    0.91
     such
    0.91
     verkar
    0.91
     такой
    0.88
    Activations Density 0.003%

    No Known Activations