© 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. 70406
    Prev
    Next
    INDEX
    Explanations

    tween and teen words

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

    the neuron activates on tokens containing the letter sequence “twe,” regardless of context (e.g. “Tween,” “tweepy,” “Tweeter”).

    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
     a
    -2.89
    }
    -2.83
    ?’
    -2.75
     ‘
    -2.64
    !”
    -2.52
    ”
    -2.42
    も同じ
    -2.34
    呖
    -2.33
     одна
    -2.33
     overarching
    -2.30
    POSITIVE LOGITS
    anyeol
    2.50
    禵
    2.45
    after
    2.39
    鹮
    2.19
    because
    2.16
    黽
    2.16
    鯪
    2.14
     ночью
    2.08
    Setelah
    2.06
    you
    2.05
    Activations Density 0.003%

    No Known Activations