© 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. 520
    Prev
    Next
    INDEX
    Explanations

    syntax elements and structures in code, particularly related to function definitions and type annotations

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

    programming-related syntax and technical code structures, particularly around type definitions, functions, and URL parsing in programming contexts.

    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
     betweenstory
    -0.81
    SequentialGroup
    -0.78
     pleaſure
    -0.65
     للاسماء
    -0.63
    MLLoader
    -0.61
    UnitTesting
    -0.60
    Sucesor
    -0.60
    DockStyle
    -0.59
     purpoſe
    -0.58
    afficheront
    -0.57
    POSITIVE LOGITS
     a
    0.49
     an
    0.46
     instanceof
    0.45
    Ljava
    0.40
     تضيفلها
    0.39
     small
    0.38
     typical
    0.38
    memberof
    0.37
     tiny
    0.36
     place
    0.35
    Activations Density 0.102%

    No Known Activations