© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99057
    Prev
    Next
    INDEX
    Explanations

    pan followed by part of word

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

    The neuron detects occurrences of the letter sequence “pan” appearing as a prefix in 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
     gelingt
    -0.91
    隙間
    -0.79
     Spani
    -0.73
     FRESH
    -0.71
     ocorreu
    -0.70
     exposed
    -0.69
     próprio
    -0.69
    彫刻
    -0.67
    gcs
    -0.67
     lẻ
    -0.66
    POSITIVE LOGITS
     pan
    1.88
    Pan
    1.80
     Pan
    1.75
    PAN
    1.59
     PAN
    1.55
    pan
    1.48
     Panther
    1.28
     Пан
    1.27
     Panthers
    1.24
     пан
    1.22
    Activations Density 0.026%

    No Known Activations