© 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. 14315
    Prev
    Next
    INDEX
    Explanations

    proper nouns, locations, and facts

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

    The neuron is specifically triggered by numeric and scientific‐notation tokens—numbers, chemical formula pieces, parentheses, measurement signs—typical of technical or scientific descriptions.

    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
    asiest
    -1.00
    iramente
    -1.00
     соответственно
    -0.99
     朋友
    -0.98
    alamu
    -0.97
     invités
    -0.96
    wanne
    -0.96
    raisemb
    -0.96
    んですよね
    -0.96
    应该
    -0.92
    POSITIVE LOGITS
     you
    1.00
    内容は
    0.99
    rinos
    0.98
    zg
    0.95
     heath
    0.94
    𝟵
    0.93
     prevented
    0.93
     with
    0.93
     diesmal
    0.90
     swamp
    0.89
    Activations Density 0.003%

    No Known Activations