© 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. 54450
    Prev
    Next
    INDEX
    Explanations

    sentence endings and new starts

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

    The neuron is highly responsive to blank‐line or paragraph‐break tokens, i.e. it detects and marks section or paragraph boundaries.

    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
     it
    -1.04
    idig
    -1.02
    словно
    -1.02
    совершен
    -1.01
    crollView
    -0.98
    viz
    -0.96
    begin
    -0.96
     starting
    -0.95
    Your
    -0.93
     namelijk
    -0.93
    POSITIVE LOGITS
     another
    1.41
     others
    1.20
     also
    1.18
     later
    1.16
     anderes
    1.05
     vieles
    1.05
     επίσης
    1.01
     آخر
    1.00
     nových
    0.99
     autre
    0.99
    Activations Density 0.009%

    No Known Activations