© 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. 22-GEMMASCOPE-RES-131K
    7. 39860
    Prev
    Next
    INDEX
    Explanations

    being lost

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

    The neuron fires on words and phrases that express something being lost, disappeared, or drowned out (e.g. “lost,” “disappear,” “swept aside,” “cloaked”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     alun
    -0.88
     originally
    -0.86
     GEN
    -0.85
    ticus
    -0.85
     แก
    -0.84
     ın
    -0.84
     Hamp
    -0.84
     POL
    -0.84
     STAN
    -0.83
     manualmente
    -0.83
    POSITIVE LOGITS
     mismas
    0.95
     ocurrir
    0.95
     detayl
    0.91
     respectivos
    0.91
     něj
    0.90
     begitu
    0.89
    irée
    0.89
    цією
    0.88
    зне
    0.85
    Anand
    0.85
    Activations Density 0.060%

    No Known Activations