© 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. 50290
    Prev
    Next
    INDEX
    Explanations

    data and sample processing

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

    The neuron fires on terms naming stages or steps of a scientific procedure (e.g. sample, preparation, pre-, post-, treatment).

    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
    娼
    -0.79
    chenko
    -0.76
    dessus
    -0.75
    katan
    -0.74
     Conquest
    -0.73
     drafting
    -0.73
     sociology
    -0.73
    šč
    -0.72
    ấp
    -0.72
     voting
    -0.72
    POSITIVE LOGITS
     processing
    3.63
     Processing
    2.94
     preprocessing
    2.88
    处理
    2.84
    Processing
    2.77
    processing
    2.70
     processed
    2.59
     PROCESSING
    2.42
    處理
    2.38
     обрабо
    2.30
    Activations Density 0.112%

    No Known Activations