© 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. 90030
    Prev
    Next
    INDEX
    Explanations

    mistakes

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

    The neuron activates on occurrences of the root “mistake” (e.g. “mistake,” “mistakes,” “mistaken”).

    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
    เพื่อ
    -0.98
     abnormal
    -0.94
    Abnormal
    -0.91
     to
    -0.90
     insane
    -0.89
    ’
    -0.88
    strange
    -0.88
     complesso
    -0.86
     realmente
    -0.86
    se
    -0.85
    POSITIVE LOGITS
     Errors
    1.81
     errors
    1.80
     mistakes
    1.75
     Mistakes
    1.66
     ошибок
    1.49
     mistake
    1.41
     errores
    1.41
     ошибки
    1.41
     erreur
    1.40
    Errors
    1.39
    Activations Density 0.031%

    No Known Activations