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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14837
    Prev
    Next
    INDEX
    Explanations

    programming-related keywords and constructs, particularly those involving data structures and methods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AccessorTable
    -0.85
     виправивши
    -0.73
    SequentialGroup
    -0.71
    TagMode
    -0.71
    expandindo
    -0.65
    migrationBuilder
    -0.64
    AddTagHelper
    -0.63
    HtmlAttribute
    -0.63
    MarshalTo
    -0.62
    ViewFeatures
    -0.62
    POSITIVE LOGITS
    ագրություններ
    0.30
    makta
    0.29
     Daily
    0.29
     TAC
    0.28
     ila
    0.27
     rhetoric
    0.26
    ,
    0.26
     EP
    0.26
    Buenas
    0.26
    kapet
    0.26
    Activations Density 1.077%

    No Known Activations