© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 14587
    Prev
    Next
    INDEX
    Explanations

    breach and caused

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

    tokens that are large numeric (floating-point) constants.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0

    the neuron detects sentence-ending or discourse boundary tokens (tokens at ends of sentences/utterances, like final punctuation and sentence-final words).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     היו
    0.51
     进行
    0.48
     подачи
    0.48
    ÑO
    0.48
    िनियम
    0.47
     periodistas
    0.47
     እንቅስቃሴ
    0.47
    胗
    0.47
     Gosudarstvennyj
    0.46
     nonEmpty
    0.46
    POSITIVE LOGITS
    an
    0.57
    ar
    0.52
    scott
    0.51
    a
    0.50
    final
    0.49
    key
    0.49
    fellow
    0.48
    ed
    0.47
     All
    0.47
    y
    0.47
    Activations Density 0.000%

    No Known Activations