© 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. 60598
    Prev
    Next
    INDEX
    Explanations

    leaving or leaving behind

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

    This neuron activates on imperative‐style action words—verbs used as commands or instructions (e.g. “mov,” “unmap,” “please,” “have,” “trim”).

    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
    ム
    -0.94
    inson
    -0.86
    轰
    -0.83
    योग
    -0.80
    เปิด
    -0.80
    क्ष
    -0.80
    व
    -0.79
    akos
    -0.78
    Opening
    -0.78
    اول
    -0.77
    POSITIVE LOGITS
     释放
    1.13
     leaving
    1.05
    releasing
    1.01
     leaves
    0.98
     katk
    0.96
    leaving
    0.95
     return
    0.94
    คืน
    0.92
     releasing
    0.92
     leave
    0.92
    Activations Density 0.002%

    No Known Activations