© 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. 34-GEMMASCOPE-RES-131K
    7. 91843
    Prev
    Next
    INDEX
    Explanations

    convention and conventional methods

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

    The neuron strongly activates on occurrences of the word “convention” (and its morphological variants, e.g. “conventionally”).

    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
    often
    -0.93
     immense
    -0.93
     console
    -0.92
     composta
    -0.91
    weird
    -0.91
     семь
    -0.89
    different
    -0.88
     adventure
    -0.88
    usually
    -0.87
    simply
    -0.86
    POSITIVE LOGITS
     conven
    1.51
     Conven
    1.47
     Conv
    1.45
     conv
    1.39
    conv
    1.38
     CONV
    1.36
    Conven
    1.34
    CONV
    1.33
     convent
    1.30
     Conventional
    1.29
    Activations Density 0.024%

    No Known Activations