© 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. 42816
    Prev
    Next
    INDEX
    Explanations

    telegram and telegrams

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

    The neuron fires on any token containing the “tele-” string (e.g. Telegram, telegrams, teleport, telegraphic).

    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
     brooding
    -2.72
    鸰
    -2.67
    岖
    -2.53
    鲠
    -2.39
    百里
    -2.39
    rógeno
    -2.38
    塬
    -2.38
     RequestMethod
    -2.38
    或者
    -2.36
     všech
    -2.36
    POSITIVE LOGITS
    ”,
    2.50
     Sicherung
    2.44
    𖥧
    2.39
    “
    2.36
    ;
    2.36
    ”;
    2.33
     acompañada
    2.30
    鬈
    2.23
     accedi
    2.23
    ቍ
    2.22
    Activations Density 0.001%

    No Known Activations