© 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. 28343
    Prev
    Next
    INDEX
    Explanations

    entertainment and purposes

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

    The neuron fires on occurrences of the word “entertainment” (and closely related headings or mentions of entertainment).

    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
    s
    -2.61
    2
    -2.53
     aperto
    -2.52
    '
    -2.36
     vuole
    -2.36
    
    -2.14
    -2.13
     ricorda
    -2.08
    D
    -2.05
    天生
    -2.03
    POSITIVE LOGITS
    .
    3.64
    涣
    2.17
     sogen
    2.17
    がありますが
    2.13
    があります
    2.13
    ྅
    2.08
     公园
    2.06
    鉝
    2.03
    </h3>
    2.02
    ってみた
    2.00
    Activations Density 0.008%

    No Known Activations