© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34579
    Prev
    Next
    INDEX
    Explanations

    comedy and satire

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

    This neuron is basically a general‐purpose “token detector” that fires moderately on almost any word and most strongly on end-of-sentence punctuation (the period).

    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
    ваясь
    -1.14
    歟
    -1.03
     چیه
    -1.02
    欤
    -0.96
    焯
    -0.96
     comprendere
    -0.95
     preliminar
    -0.94
     yesterday
    -0.94
    izacin
    -0.94
     heartbreaking
    -0.94
    POSITIVE LOGITS
     comedy
    1.49
     hilarious
    1.30
     divertidas
    1.24
     humor
    1.24
     humour
    1.24
     comedic
    1.23
     divertidos
    1.18
    laag
    1.17
    praia
    1.08
     alkoholu
    1.08
    Activations Density 0.183%

    No Known Activations