© 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. 42206
    Prev
    Next
    INDEX
    Explanations

    opinion, new, foreign characters

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

    The neuron fires on informal discourse markers and hedging phrases (e.g. “Anyway,” “Actually,” “Like,” “I think,” “Idk,” “Also,” “Okay,” etc.) that structure casual conversation.

    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
     +
    -1.73
    "
    -1.69
     to
    -1.58
     “
    -1.53
     https
    -1.52
    なら
    -1.50
    !
    -1.50
     data
    -1.50
     размера
    -1.49
     we
    -1.48
    POSITIVE LOGITS
     Eſ
    1.90
     TÉCNICA
    1.72
    Forrás
    1.63
     дів
    1.61
     (~(
    1.60
     éstos
    1.60
    ㍑
    1.60
     chré
    1.59
     silang
    1.57
    】:
    1.55
    Activations Density 0.648%

    No Known Activations