© 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. 65399
    Prev
    Next
    INDEX
    Explanations

    different approach

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

    This neuron detects occurrences of the word “different” (and closely related comparative constructions).

    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
    躂
    -1.63
    茀
    -1.62
    瑒
    -1.59
    ዣ
    -1.56
    曖昧さ
    -1.55
     Даже
    -1.52
    ⵈ
    -1.49
    逡
    -1.49
    ſtra
    -1.48
    Решение
    -1.46
    POSITIVE LOGITS
     than
    3.13
     they
    1.79
     we
    1.70
    hasMoreElements
    1.66
     know
    1.66
     felt
    1.66
    tım
    1.65
     затем
    1.62
     []:
    1.59
     then
    1.56
    Activations Density 0.057%

    No Known Activations