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 - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 955
    Prev
    Next
    INDEX
    Explanations

    code-related constructs, including function definitions and array operations

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

    Code snippets including "if"

    np_acts-logits-general · gemini-2.0-flash

    conditional programming statements or phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     initComponents
    -0.50
     eth
    -0.47
    Camping
    -0.41
    phazard
    -0.40
    VersionUID
    -0.40
    udic
    -0.39
     CallOverrides
    -0.39
     barnen
    -0.39
    sune
    -0.38
    ombic
    -0.38
    POSITIVE LOGITS
    if
    0.47
     <=",
    0.46
    ViewImports
    0.44
     дописавши
    0.44
    GOTREF
    0.42
     للاسماء
    0.42
    ExtendWith
    0.42
    tvguidetime
    0.41
    SOUNDBITE
    0.41
    elemField
    0.41
    Activations Density 0.168%

    No Known Activations