© 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. 68569
    Prev
    Next
    INDEX
    Explanations

    demonstrating by comparison

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

    The neuron detects discourse cues that introduce justification or evidence (e.g. “by,” “considering,” “illustrated,” “look,” “need”).

    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
    matory
    -1.12
    teness
    -0.97
    uding
    -0.92
    uities
    -0.91
     это
    -0.87
    х
    -0.86
    inglés
    -0.86
    ulation
    -0.85
     głównie
    -0.84
    伐
    -0.84
    POSITIVE LOGITS
     when
    1.55
     comparison
    1.34
     Tatsache
    1.29
     Betrachtung
    1.10
     recent
    1.09
     ouvir
    1.02
     Saludos
    1.01
     compared
    1.00
     lorsque
    0.98
     Após
    0.98
    Activations Density 0.042%

    No Known Activations