© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 7194
    Prev
    Next
    INDEX
    Explanations

    occurrences of symbols and special characters commonly used in coding or technical documents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    code or technical file paths with programming syntax elements.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    technical software-engineering artifacts in text, especially source-code/diff snippets, stack traces, file paths, and low-level encoded identifiers

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    numeric characters in source code, particularly in file paths, version numbers, and reference identifiers.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dolayı
    -0.37
    aarrggbb
    -0.36
     Füßen
    -0.35
    AutoScaleMode
    -0.34
     Wikiseite
    -0.34
     ardından
    -0.33
    AndroidJUnit
    -0.32
     Anschließend
    -0.32
     चुनें
    -0.32
     aislado
    -0.31
    POSITIVE LOGITS
    ſem
    0.60
    رشف
    0.58
    ſchen
    0.57
     OneNote
    0.56
    клопе
    0.56
     Undertaker
    0.55
    ymce
    0.55
    왘
    0.53
    parametrize
    0.53
    Gön
    0.52
    Activations Density 0.050%

    No Known Activations