© 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. 10-GEMMASCOPE-RES-131K
    7. 65451
    Prev
    Next
    INDEX
    Explanations

    routes and directions

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

    The neuron fires on the heading “Directions” (especially when followed by “to”), i.e., section titles that introduce route or navigation instructions.

    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
    ࿙
    -2.23
    孭
    -2.22
     Weeknd
    -2.08
     вернулся
    -2.08
     работал
    -2.08
     принял
    -2.06
    睒
    -2.03
    騭
    -2.03
     работе
    -2.02
    釹
    -2.00
    POSITIVE LOGITS
     Our
    3.63
    .
    2.73
    ↵↵
    2.53
     directions
    2.52
     unsere
    2.42
     is
    2.31
    抔
    2.08
     verbess
    2.02
     Nachweis
    1.98
     Unsere
    1.97
    Activations Density 0.004%

    No Known Activations