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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1030
    Prev
    Next
    INDEX
    Explanations

    instances of exception handling and error reporting in programming contexts

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

    coding syntax related to error handling and exceptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     AspNetCore
    -0.89
    webElementXpaths
    -0.85
    RenderAtEndOf
    -0.84
    舺
    -0.77
    SharedCtor
    -0.77
    UrlResolution
    -0.77
    intios
    -0.77
    RefNanny
    -0.75
    DebuggerNonUser
    -0.73
    出版年
    -0.72
    POSITIVE LOGITS
    ruptedException
    0.55
     Anfänger
    0.51
     suivantes
    0.47
     Hitam
    0.47
     subsequent
    0.47
     seura
    0.47
    >>()
    0.45
     following
    0.44
     problems
    0.44
     unexpected
    0.42
    Activations Density 0.003%

    No Known Activations