© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 432
    Prev
    Next
    INDEX
    Explanations

    subjective judgments

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

    The neuron fires on subjective, evaluative adjectives conveying feelings or judgments (e.g. “strange,” “intriguing,” “frustrating,” “obnoxious,” “possible”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     lucru
    -1.20
    lettre
    -1.19
    embang
    -1.17
     singur
    -1.16
     autén
    -1.16
     desconocido
    -1.15
     merkwür
    -1.15
     acesta
    -1.15
     nimic
    -1.15
     attribué
    -1.14
    POSITIVE LOGITS
     to
    3.14
     that
    2.00
     because
    1.77
     for
    1.67
     when
    1.52
     you
    1.28
     since
    1.20
     how
    1.17
     these
    1.16
     people
    1.16
    Activations Density 0.052%

    No Known Activations