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. 2468
    Prev
    Next
    INDEX
    Explanations

    financial metrics and statistical values related to company performance

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

    mathematical calculations and expressions

    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
     defaultstate
    -0.81
    Rhestr
    -0.74
    FieldBuilder
    -0.73
    PMailer
    -0.73
    RenderAtEndOf
    -0.73
    Hentet
    -0.72
     Photocase
    -0.71
    Cyfarwyddwr
    -0.68
    aarrggbb
    -0.68
    httphttps
    -0.67
    POSITIVE LOGITS
     dolay
    0.32
    bilecek
    0.28
     wrong
    0.26
     forhold
    0.26
     Zahlungs
    0.26
     perdiendo
    0.24
     willing
    0.23
    ebenarnya
    0.23
     bomberos
    0.23
     necesitas
    0.23
    Activations Density 0.052%

    No Known Activations