© 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. 54328
    Prev
    Next
    INDEX
    Explanations

    low dose and concentrations

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

    The neuron detects the word "low" often in scientific or medical contexts, particularly when describing measurements, dosages, temperatures, or rates.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    ?”
    -2.80
    >
    -2.52
      
    -2.33
     vorigen
    -2.23
     generando
    -2.20
    In
    -2.17
     “(
    -2.16
    曁
    -2.16
    跄
    -2.16
     dritten
    -2.09
    POSITIVE LOGITS
    fying
    2.45
     られる
    2.39
     chuť
    2.33
    玬
    2.31
    ні
    2.23
    s
    2.23
    ченные
    2.19
     problemen
    2.19
    妗
    2.19
     με
    2.17
    Activations Density 0.046%

    No Known Activations