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-2-9B
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 129896
    Prev
    Next
    INDEX
    Explanations

    specific programming syntax related to object creation and parsing functions

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

    section identifiers using {#}

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     snippetHide
    -0.67
    onViewCreated
    -0.63
    MigrationBuilder
    -0.61
     ujednoznacz
    -0.61
    CppMethod
    -0.58
     الحره
    -0.56
     Grüsse
    -0.54
     برانيه
    -0.54
    MessageTagHelper
    -0.54
    IntoConstraints
    -0.54
    POSITIVE LOGITS
    Vidite
    0.37
     PyObject
    0.36
     "
    0.35
    readystatechange
    0.34
    essentiel
    0.32
    ਿ
    0.31
    に乗って
    0.31
    所谓
    0.31
    x
    0.31
     htons
    0.30
    Activations Density 0.227%

    No Known Activations