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. 91206
    Prev
    Next
    INDEX
    Explanations

    significant contextual markers or tags in structured data

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

    context before user

    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
     Kā
    -0.38
     Mo
    -0.38
     Sel
    -0.36
     Ad
    -0.36
     Spin
    -0.35
     spin
    -0.35
    spin
    -0.35
    ysio
    -0.35
     mo
    -0.34
     Zak
    -0.34
    POSITIVE LOGITS
    uxxxx
    0.72
     Italijanski
    0.71
    Personensuche
    0.65
     Taktlose
    0.63
    enderror
    0.62
    TagMode
    0.61
    ViewImports
    0.60
    tagHelperRunner
    0.59
    tonode
    0.59
    StandardCharsets
    0.58
    Activations Density 0.000%

    No Known Activations