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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 5198
    Prev
    Next
    INDEX
    Explanations

    punctuation and syntax-related elements in programming code

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

    code import statements from various programming languages like JavaScript, Python, and C.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    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
     للاسماء
    -0.82
    
    -0.78
    AndEndTag
    -0.76
     فريبيس
    -0.74
    RTLD
    -0.73
    aarrggbb
    -0.73
    StructEnd
    -0.71
    ftagPool
    -0.71
    TagMode
    -0.71
    StoreMessageInfo
    -0.70
    POSITIVE LOGITS
     and
    0.39
     trône
    0.34
     lendemain
    0.33
    พาะ
    0.31
     equipos
    0.31
     dedans
    0.30
     fondateur
    0.29
     podjet
    0.29
    INTERVAL
    0.29
     Setting
    0.29
    Activations Density 0.153%

    No Known Activations