© 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. 83886
    Prev
    Next
    INDEX
    Explanations

    correcting or compensating for errors

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

    The neuron flags technical‐method tokens in scientific or patent‐style text—especially words related to correction/compensation procedures (e.g. resistance, correction, calibration, compensation, maintained, after).

    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
    cznych
    -0.88
     katanya
    -0.79
     bezpiecze
    -0.79
    easing
    -0.78
    ствовал
    -0.77
    cił
    -0.77
     sắp
    -0.76
     торт
    -0.75
     Schicks
    -0.72
    ]>=
    -0.72
    POSITIVE LOGITS
     correction
    3.69
     corrections
    3.27
     Correction
    2.89
    correction
    2.81
    Correction
    2.81
     correcting
    2.69
     corrected
    2.63
     Corrections
    2.58
     corrects
    2.52
    corrected
    2.30
    Activations Density 0.088%

    No Known Activations