© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44348
    Prev
    Next
    INDEX
    Explanations

    here is/are

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

    The neuron strongly activates on the word “here” (in both lower- and upper-case) when used as a discourse marker (e.g. “Here is…” or “Here are…”).

    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
     the
    -1.76
     canzoni
    -1.57
     ストラップ
    -1.47
     vru
    -1.47
     daer
    -1.47
     prins
    -1.46
     ボール
    -1.45
     groet
    -1.44
     définitivement
    -1.41
     staffel
    -1.41
    POSITIVE LOGITS
     WANT
    1.65
     GREAT
    1.63
    谘
    1.62
     TWO
    1.58
    Through
    1.57
    During
    1.56
     paredes
    1.52
     WITH
    1.48
     primeras
    1.48
    Instead
    1.47
    Activations Density 0.027%

    No Known Activations