© 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. 34-GEMMASCOPE-RES-131K
    7. 93495
    Prev
    Next
    INDEX
    Explanations

    fus, fuse, fusion

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

    The neuron detects forms of the word "fusion" or "fus-" prefixes in text, particularly in technical contexts like programming, science, or legal documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     crisi
    -0.94
    无语
    -0.89
    Dough
    -0.86
     peliculas
    -0.85
     znale
    -0.84
    氈
    -0.82
     humaines
    -0.82
    dateOfBirth
    -0.82
     argento
    -0.81
     fiore
    -0.80
    POSITIVE LOGITS
     fus
    1.44
     Fus
    1.22
    Fus
    1.20
     Fuse
    1.08
     fuse
    1.08
     fusi
    1.02
    fuse
    1.02
    fus
    1.02
     Fusion
    1.01
    Fuse
    1.00
    Activations Density 0.029%

    No Known Activations