© 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. 22-GEMMASCOPE-RES-131K
    7. 9321
    Prev
    Next
    INDEX
    Explanations

    no effect or unimportant

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

    This neuron responds to contrastive or exception-marking discourse cues—words like “but,” “however,” “when,” and related phrases that introduce a contrasting or exception clause.

    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
    '],
    -0.87
     estrateg
    -0.84
    (_.
    -0.81
     FirebaseAuth
    -0.79
     レシピ
    -0.78
     stie
    -0.77
     kemenangan
    -0.77
    ducción
    -0.77
     abiti
    -0.77
     вместо
    -0.77
    POSITIVE LOGITS
     negligible
    1.45
     unimportant
    1.40
     insignificant
    1.38
     unlikely
    1.35
     minor
    1.32
     immaterial
    1.30
     irrelevant
    1.29
     insign
    1.27
     incon
    1.19
    不会
    1.14
    Activations Density 0.038%

    No Known Activations