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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5747
    Prev
    Next
    INDEX
    Explanations

    programming concepts related to linking and dependencies in software development

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

    setting libraries or options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    -0.71
    脚注の使い方
    -0.57
    sego
    -0.50
    TypedDataSet
    -0.50
    Portály
    -0.50
    protoimpl
    -0.50
    contrar
    -0.50
    توض
    -0.49
    buttonBar
    -0.49
     ilustração
    -0.48
    POSITIVE LOGITS
    Obrázky
    0.37
    IntoConstraints
    0.32
    :+:
    0.32
     defaultstate
    0.31
     initComponents
    0.30
    企
    0.30
    Tasting
    0.29
    setHorizontal
    0.29
     Ü
    0.29
     stage
    0.29
    Activations Density 0.082%

    No Known Activations