© 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. 22-GEMMASCOPE-RES-131K
    7. 12897
    Prev
    Next
    INDEX
    Explanations

    game outcomes

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

    The neuron is keyed to spotting combat‐sports finishers—especially grappling submission phrases like “rear‐naked choke for the submission victory.”

    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
     these
    -1.13
    THEY
    -1.07
     fabriqué
    -1.06
    But
    -1.03
     этих
    -1.02
    irci
    -1.02
    Its
    -0.98
    当たり前
    -0.98
    zőkönyv
    -0.96
     riservata
    -0.96
    POSITIVE LOGITS
     just
    1.02
     gave
    0.98
     instantly
    0.98
     without
    0.97
     second
    0.92
    шње
    0.91
    Viri
    0.90
     took
    0.89
     motivo
    0.89
     پری
    0.88
    Activations Density 0.008%

    No Known Activations