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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1279
    Prev
    Next
    INDEX
    Explanations

    patterns and codes related to data structures or programming elements

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

    Appears to be extracting numbers

    np_acts-logits-general · gemini-2.0-flash

    hex identifiers and numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
     Administrativna
    -0.94
     Italijanski
    -0.71
    styleType
    -0.69
     OFDb
    -0.69
    Autoritní
    -0.66
    
    -0.65
     AssemblyProduct
    -0.64
    ftagPool
    -0.61
    LookAnd
    -0.59
     kaarangay
    -0.59
    POSITIVE LOGITS
     within
    0.34
     whole
    0.34
     gusta
    0.30
    DISPATCH
    0.30
     this
    0.30
     mes
    0.30
    Readable
    0.30
     view
    0.29
     aspect
    0.29
     under
    0.29
    Activations Density 0.340%

    No Known Activations