© 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. 10-GEMMASCOPE-RES-131K
    7. 65010
    Prev
    Next
    INDEX
    Explanations

    Pelican Bottles, AMC

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

    The neuron fires on fragments of proper names—specifically on tokens from brand or technical names like “Pelican” and “Wadsworth.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ︐
    -3.17
     mesmos
    -2.89
    饜
    -2.88
     vantagens
    -2.84
    ⸰
    -2.61
    笮
    -2.59
    縢
    -2.58
     jezt
    -2.58
     bidra
    -2.56
    鼱
    -2.53
    POSITIVE LOGITS
     ingenious
    3.23
     exquisite
    3.02
     exquisitely
    2.88
     burgeoning
    2.78
     poignant
    2.73
     jovial
    2.66
     phenomenal
    2.59
     exuberant
    2.59
    ↵↵
    2.56
    ’.
    2.56
    Activations Density 0.015%

    No Known Activations