© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 118329
    Prev
    Next
    INDEX
    Explanations

    titles and names

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

    The neuron strongly activates on unusually long tokens (e.g. multi‐syllable technical terms), effectively detecting long words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     longed
    -0.79
     LENGTH
    -0.78
     pancake
    -0.77
     Unesco
    -0.77
    Inte
    -0.75
     littoral
    -0.70
    especial
    -0.70
     Pancake
    -0.70
     威
    -0.70
    队列
    -0.69
    POSITIVE LOGITS
     event
    0.91
     Licensing
    0.82
    ртка
    0.80
     licensing
    0.77
    VP
    0.77
     Event
    0.75
    IP
    0.74
    calibration
    0.74
    MessageHandler
    0.74
     IP
    0.73
    Activations Density 0.018%

    No Known Activations