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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 2493
    Prev
    Next
    INDEX
    Explanations

    terms related to actionable insights and decision-making in various contexts

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

    actionable decisions and actions

    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
     reír
    -0.36
     senhora
    -0.34
     Erweiter
    -0.32
     EXCELL
    -0.32
     cielos
    -0.30
     excelencia
    -0.30
    gridad
    -0.30
    ürü
    -0.30
     satisfecho
    -0.30
    赔
    -0.29
    POSITIVE LOGITS
     utafitiHapana
    0.72
     فريبيس
    0.69
    Diweddarwch
    0.63
    Jereo
    0.60
    HandlerContext
    0.60
    ьаж
    0.60
     صوتيه
    0.60
    bootstrapcdn
    0.59
     @"/
    0.59
    DispatchToProps
    0.58
    Activations Density 0.099%

    No Known Activations