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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28452
    Prev
    Next
    INDEX
    Explanations

    LaTeX syntax and mathematical notation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     DriverManager
    -0.86
    writerow
    -0.79
    hnia
    -0.76
    creativecommons
    -0.75
    currentColor
    -0.74
    rosophila
    -0.73
    ChildScrollView
    -0.70
     ويكيميديا
    -0.70
    chyma
    -0.68
    alom
    -0.68
    POSITIVE LOGITS
     boc
    0.63
     indeed
    0.60
     Gos
    0.59
    indeed
    0.57
     olev
    0.55
    SUB
    0.52
     Bellow
    0.51
    Datuak
    0.51
     Teles
    0.50
    mortem
    0.50
    Activations Density 0.064%

    No Known Activations