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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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 - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3199
    Prev
    Next
    INDEX
    Explanations

    occurrences of constructor and initialization function definitions in programming code

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

    "constructor" and related terms

    np_acts-logits-general · gemini-2.0-flash

    constructor and init methods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     @"/
    -0.67
     "..\..\
    -0.59
    (&:
    -0.57
     QtCore
    -0.56
     "../../../../
    -0.53
    aarrggbb
    -0.53
    ifornie
    -0.53
    +#+#
    -0.53
    migrationBuilder
    -0.53
    modelBuilder
    -0.49
    POSITIVE LOGITS
    constructor
    0.92
    Bronnen
    0.72
    ArgsConstructor
    0.70
     constructor
    0.69
    tagHelperRunner
    0.66
    曖昧さ回避
    0.66
    HtmlAttribute
    0.63
     ProtoMessage
    0.60
     构造
    0.55
     bols
    0.54
    Activations Density 0.021%

    No Known Activations