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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-GEMMASCOPE-TRANSCODER-16K
    4. 2124
    Prev
    Next
    INDEX
    Explanations

    a mixture of formatting symbols and fractional numbers with sporadic words or abbreviations from different topics such as gaming or programming

    oai_token-act-pair · gemini-2.0-flash

    foreign words

    np_max-act-logits · gemini-2.0-flash

    Common code patterns for a boss AI in a game, including timers, spell casting, and instance management. Uses pointer syntax and specific naming conventions for game-related functions and variables.

    eleuther_acts_top20 · claude-3-5-sonnet-20240620Triggered by @oreo4455

    code and programming syntax elements, particularly in video game programming.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @oreo4455
    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
    Autoritní
    -0.53
     propOrder
    -0.52
    ніципалі
    -0.51
     चीज़ों
    -0.49
    CppCodeGen
    -0.49
    Amicalement
    -0.48
     viewDidLoad
    -0.47
    bability
    -0.47
     Autorizaciones
    -0.46
     ModelExpression
    -0.45
    POSITIVE LOGITS
     кӀ
    0.52
    исленность
    0.52
    assertj
    0.52
    ServletException
    0.49
    тьяна
    0.49
    ffie
    0.48
    tip
    0.47
    iolis
    0.47
    providedIn
    0.46
     enfrentan
    0.45
    Activations Density 0.229%

    No Known Activations