© 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. 10-GEMMASCOPE-RES-131K
    4. 21656
    Prev
    Next
    INDEX
    Explanations

    related to things followed by certain connectors

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

    The neuron fires on past-participle or passive-voice verbs (e.g. “celebrated,” “conducted,” “followed,” “dibuat oleh”) that mark actions or events.

    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
    臯
    -1.84
    使った
    -1.80
     \
    -1.77
     acorde
    -1.73
    一个月
    -1.73
     
    -1.72
    ではある
    -1.68
     entreten
    -1.65
     claros
    -1.63
     vistos
    -1.62
    POSITIVE LOGITS
     by
    2.73
     These
    2.31
    猡
    2.14
    There
    2.08
    These
    1.92
     चीज़ों
    1.84
     Putih
    1.82
     because
    1.82
    。『
    1.82
     istimewa
    1.79
    Activations Density 0.001%

    No Known Activations