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. 8176
    Prev
    Next
    INDEX
    Explanations

    sentences containing quotes and dialogue markers

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

    The token "Q:" or "Q" marks the beginning of questions in structured dialogue formats, such as court transcripts, interviews, or online Q&A forums.

    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
    évaluateur
    -0.60
    DIPSETTING
    -0.60
    UrlResolution
    -0.56
    IsMutable
    -0.53
    PerformLayout
    -0.52
     pinulongan
    -0.51
    ✭✭
    -0.50
    sizeCache
    -0.50
     ویکی‌پدی
    -0.49
     AssemblyCompany
    -0.48
    POSITIVE LOGITS
    enterOuterAlt
    0.40
    ">//
    0.38
     '{@
    0.38
     CommonModule
    0.37
     svolta
    0.33
    atschappij
    0.32
     StatelessWidget
    0.32
    stdafx
    0.30
     getopt
    0.30
    怎么办
    0.30
    Activations Density 0.065%

    No Known Activations