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. 20-GEMMASCOPE-TRANSCODER-16K
    4. 14080
    Prev
    Next
    INDEX
    Explanations

    hexadecimal numbers, specific words "Lane" and "three", and numbers encased in parenthesis

    oai_token-act-pair · gemini-2.0-flash

    Code identifiers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/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
     kedua
    -0.91
     two
    -0.89
     Kedua
    -0.84
     Two
    -0.83
     deux
    -0.81
    2
    -0.81
    two
    -0.80
     Second
    -0.79
     second
    -0.79
    二
    -0.77
    POSITIVE LOGITS
    UnusedPrivate
    0.86
    rrggbb
    0.76
     thrice
    0.71
    ❹
    0.66
    Organisateur
    0.65
    хьтан
    0.62
    reszcie
    0.61
    SerializedSize
    0.61
     quadruple
    0.61
     triple
    0.60
    Activations Density 3.222%

    No Known Activations