© 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. 28443
    Prev
    Next
    INDEX
    Explanations

    in/on/to/of something

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

    The neuron fires strongly on the first content words at the start of new text segments (e.g. sentence or paragraph beginnings).

    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
    奬
    -1.70
     निम्नलिखित
    -1.30
      
    -1.30
     lópez
    -1.29
     garcía
    -1.28
     gonzález
    -1.27
    PRWEB
    -1.27
    謚
    -1.25
    攥
    -1.23
     PHILADELPHIA
    -1.22
    POSITIVE LOGITS
     things
    1.77
     it
    1.55
    Kenapa
    1.52
     etc
    1.52
     my
    1.46
    bàn
    1.46
     mainly
    1.37
    PUESTA
    1.36
    joven
    1.35
    Kalau
    1.33
    Activations Density 0.031%

    No Known Activations