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. 805
    Prev
    Next
    INDEX
    Explanations

    code-related error handling mechanisms

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

    instances of error handling in code, specifically related to exceptions

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

    exception handling in code

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    parsedMessage
    -0.74
     Réponses
    -0.71
    SharedCtor
    -0.68
    ंदीखरीदारी
    -0.67
    +#+#
    -0.66
     <>",
    -0.64
    LookAnd
    -0.64
     مشين
    -0.63
    ########.
    -0.63
     noDo
    -0.63
    POSITIVE LOGITS
     AttributeSet
    0.59
     PhpStorm
    0.54
     الرخصة
    0.53
    ztat
    0.52
     termica
    0.51
    UserScript
    0.50
     spirituale
    0.48
    OutputType
    0.46
    bní
    0.46
    ïsme
    0.46
    Activations Density 0.015%

    No Known Activations