© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71960
    Prev
    Next
    INDEX
    Explanations

    first posts and introductions

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

    The neuron fires on typical forum‐introduction and greeting phrases (e.g. “first post,” “new here,” “welcome to the board/forum,” “hello,” etc.).

    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
     детей
    -1.15
    仍然
    -1.09
     again
    -1.05
     igjen
    -1.03
    étés
    -0.98
     mungkin
    -0.98
     still
    -0.96
     todavía
    -0.96
     actionTypes
    -0.96
     Again
    -0.95
    POSITIVE LOGITS
     joining
    1.43
     introductions
    1.26
     join
    1.23
    joining
    1.18
     welcome
    1.18
     Joining
    1.16
     Welcome
    1.13
     bienven
    1.13
    Joining
    1.08
    Welcome
    1.06
    Activations Density 0.072%

    No Known Activations