© 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. 34-GEMMASCOPE-RES-131K
    7. 38119
    Prev
    Next
    INDEX
    Explanations

    different people/things

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

    The neuron is strongly activated by very common, high-frequency “function” words (i.e. stop‐words such as “the,” “of,” “and,” 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_34/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
     jonka
    -1.45
     our
    -1.38
     nossas
    -1.30
     μας
    -1.30
     unserer
    -1.27
    ࣪
    -1.27
    我们
    -1.24
     impra
    -1.20
     mètres
    -1.20
    เรา
    -1.20
    POSITIVE LOGITS
     things
    1.77
     different
    1.73
     people
    1.59
     everything
    1.58
     particular
    1.47
     certain
    1.39
     everybody
    1.34
     everyone
    1.34
     there
    1.30
     pihaknya
    1.30
    Activations Density 0.019%

    No Known Activations