© 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. 39055
    Prev
    Next
    INDEX
    Explanations

    not need to

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

    The neuron strongly fires on the future‐passive construction—particularly the phrase “it will be done automatically” (i.e. “will be done” plus an adverb like “automatically”).

    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
    只能
    -1.02
    termilk
    -0.98
    aking
    -0.97
    andten
    -0.87
    不得
    -0.85
    rying
    -0.84
     naranja
    -0.82
     Buckley
    -0.81
    riter
    -0.81
     чехол
    -0.81
    POSITIVE LOGITS
    只需要
    1.70
    即可
    1.36
    不需要
    1.34
    就可以
    1.33
     simply
    1.23
    只需
    1.22
    只要
    1.20
     should
    1.20
    むしろ
    1.14
    就能
    1.11
    Activations Density 0.195%

    No Known Activations