© 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. 34-GEMMASCOPE-RES-131K
    7. 38255
    Prev
    Next
    INDEX
    Explanations

    verbs of action/process

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

    The neuron strongly responds to Chinese single‐character action verbs—especially 拖 (“drag”) and 滑 (“slide”)—typically describing dragging/sliding interactions.

    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
     вы
    -0.79
    丹药
    -0.78
     можно
    -0.78
     utilizes
    -0.76
     में
    -0.76
     στην
    -0.75
     impossível
    -0.73
     चीज़
    -0.73
    ari
    -0.72
    ずつ
    -0.72
    POSITIVE LOGITS
    yczą
    0.86
    寻
    0.85
    取
    0.84
    Remark
    0.77
    述
    0.76
     épreuves
    0.75
    smallskip
    0.74
    Especially
    0.74
     ज्यादा
    0.73
    QUOTES
    0.73
    Activations Density 0.005%

    No Known Activations