© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7989
    Prev
    Next
    INDEX
    Explanations

    free from defects or disease

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

    phrases that indicate something is "free of" or "free from" a negative quality or condition.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is tuned to detect the word “free” (especially in “free of…” or “free shipping/charge” contexts).

    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
    瓩
    -1.42
    tyfik
    -1.33
     ressort
    -1.33
    漚
    -1.32
     prénom
    -1.32
     attirer
    -1.30
     apporter
    -1.30
     peau
    -1.29
    只见
    -1.28
     incendie
    -1.28
    POSITIVE LOGITS
    kemeja
    1.51
    botines
    1.46
    Excelente
    1.43
    dapat
    1.43
    Terima
    1.38
     estuvo
    1.38
    resep
    1.35
    when
    1.35
     hubo
    1.34
    Visión
    1.33
    Activations Density 0.062%

    No Known Activations