© 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. 13095
    Prev
    Next
    INDEX
    Explanations

    end punctuation

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

    This neuron fires on expressions of positive feedback or thanks—e.g. “Thanks,” “thank you,” and thumbs-up/approval tokens.

    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
    autoconfigure
    -1.05
     alleged
    -1.02
     about
    -1.01
     its
    -1.01
     føl
    -0.98
     around
    -0.98
     около
    -0.98
     ähnlichen
    -0.94
     is
    -0.94
     atriz
    -0.93
    POSITIVE LOGITS
    Lastly
    1.41
    bringing
    1.16
    Paket
    1.13
    commented
    1.13
    Again
    1.13
    again
    1.13
     lastly
    1.12
     offrant
    1.09
     ekolog
    1.09
     opět
    1.08
    Activations Density 0.022%

    No Known Activations