© 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. 68334
    Prev
    Next
    INDEX
    Explanations

    specifying type or kind

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

    The neuron activates on words that introduce choices, conditions, or qualifiers—e.g. “any,” “no,” “whether,” “which,” “depending,” and “what kind of.”

    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
    などの
    -1.12
    鈎
    -1.09
    こんばん
    -1.05
    Beij
    -0.98
    Voto
    -0.96
    genome
    -0.95
    こんばんは
    -0.94
    César
    -0.94
    PHONY
    -0.93
    äumt
    -0.93
    POSITIVE LOGITS
     type
    1.70
     types
    1.55
     of
    1.41
     location
    1.41
     species
    1.37
     region
    1.30
     year
    1.20
     kind
    1.17
     genre
    1.16
     style
    1.12
    Activations Density 0.119%

    No Known Activations