Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 25-GEMMASCOPE-TRANSCODER-16K
    4. 1701
    Prev
    Next
    INDEX
    Explanations

    angle brackets used in HTML and XML, programming language syntax with parenthesis, numbers, and variables

    oai_token-act-pair · gemini-2.0-flash

    Code/script fragments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_25/width_16k
    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
    latency
    -0.52
    calibration
    -0.52
    affinity
    -0.51
    Referências
    -0.51
    complexity
    -0.47
    conductivity
    -0.47
     Barrera
    -0.47
    contamination
    -0.46
    degradation
    -0.46
    decomposition
    -0.46
    POSITIVE LOGITS
     marinho
    0.60
     $_"
    0.59
    findpost
    0.58
     RISERV
    0.55
     isNameExpr
    0.54
     crdi
    0.54
     CreateTagHelper
    0.54
     avoient
    0.54
    tvguidetime
    0.53
     venons
    0.53
    Activations Density 9.166%

    No Known Activations