© 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. 10-GEMMASCOPE-RES-131K
    7. 68924
    Prev
    Next
    INDEX
    Explanations

    sounds and interjections

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

    The neuron activates on nonlexical vocalizations or onomatopoeic tokens (e.g. “umm,” “uh,” “beep,” “da,” etc.).

    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
     and
    -1.69
    1
    -1.33
     United
    -1.29
     differences
    -1.24
     benefits
    -1.19
    gesture
    -1.18
    eyes
    -1.17
     Previous
    -1.16
     but
    -1.13
     amounts
    -1.13
    POSITIVE LOGITS
    饜
    1.48
     GRATU
    1.41
     raiſ
    1.34
    ziona
    1.31
     incont
    1.30
     uſed
    1.30
     exces
    1.30
     gravi
    1.29
     guer
    1.28
    Наши
    1.28
    Activations Density 0.028%

    No Known Activations