© 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. 10-GEMMASCOPE-RES-131K
    4. 119591
    Prev
    Next
    INDEX
    Explanations

    try:

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

    The neuron strongly fires on exception‐handling keywords and their associated clauses (e.g. “try,” “except,” “else”).

    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
     должна
    -2.20
    ate
    -2.09
    ong
    -1.93
     étant
    -1.92
    ang
    -1.91
    m
    -1.87
     Importantly
    -1.87
     elizabeth
    -1.83
    t
    -1.80
     uuidv
    -1.80
    POSITIVE LOGITS
     it
    2.23
    ẨM
    2.08
     they
    2.03
    acaktır
    1.94
    when
    1.90
    BIÉN
    1.87
    lsz
    1.86
    嗚
    1.84
     its
    1.84
    hamilan
    1.83
    Activations Density 0.002%

    No Known Activations