© 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. 109349
    Prev
    Next
    INDEX
    Explanations

    Document previews and introductions

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

    The neuron detects structural elements of slide‐deck presentations—e.g. slide titles, transcript labels, slide numbers, and their punctuation markers.

    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
     Neuro
    -0.85
    👦
    -0.84
     DbContext
    -0.79
     Cancel
    -0.79
     Erik
    -0.77
     więcej
    -0.77
    🏠
    -0.77
     плюс
    -0.77
    hawk
    -0.76
     sosyal
    -0.76
    POSITIVE LOGITS
     utilizarse
    0.84
    擠
    0.81
     slides
    0.78
     Hæ
    0.78
     disfrute
    0.77
     eficacia
    0.76
     marinas
    0.76
     unlaw
    0.76
     Fußballspieler
    0.75
     realizan
    0.75
    Activations Density 0.005%

    No Known Activations