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. 24195
    Prev
    Next
    INDEX
    Explanations

    opening curly braces in code snippets

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

    No MAX_ACTIVATING_TOKENS provided

    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
    GEBURTS
    -0.80
     виправивши
    -0.74
    richTextPanel
    -0.70
     AssemblyCompany
    -0.69
    ValueStyle
    -0.68
    دانشنامهٔ
    -0.66
    下载附件
    -0.65
     invokingState
    -0.64
    tanleria
    -0.64
     ModelExpression
    -0.62
    POSITIVE LOGITS
    PostMapping
    0.34
     claim
    0.33
     mechanic
    0.32
     Méndez
    0.32
     check
    0.31
     Jiménez
    0.31
     tip
    0.31
    colorPrimary
    0.30
     Baker
    0.30
    を確認
    0.30
    Activations Density 0.003%

    No Known Activations