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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 117392
    Prev
    Next
    INDEX
    Explanations

    assertions of personal beliefs or opinions

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

    Special characters and formatting elements that denote structure in documents, including punctuation marks, paragraph symbols, reference indicators, and HTML/code elements.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     initComponents
    -0.41
    renço
    -0.38
    toir
    -0.38
    MENAFN
    -0.37
    arangay
    -0.36
    chränk
    -0.35
    leſs
    -0.35
    íncipe
    -0.35
     VIDEOTAPE
    -0.34
    Климат
    -0.33
    POSITIVE LOGITS
    Hentet
    0.62
     informée
    0.60
     مشين
    0.47
    хьтан
    0.47
     فريبيس
    0.45
    WithIOException
    0.44
     plutôt
    0.43
     ErrIntOverflow
    0.43
    WriteBarrier
    0.42
    avadoc
    0.42
    Activations Density 0.015%

    No Known Activations