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. 14119
    Prev
    Next
    INDEX
    Explanations

    components of programming structure, particularly indentation and syntax elements

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

    code definitions and loops

    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
     pleaſure
    -0.59
    ſelf
    -0.56
     faſt
    -0.55
    outWeight
    -0.54
     tranſ
    -0.54
     raiſ
    -0.52
    HasColumnName
    -0.51
     GoogleFonts
    -0.50
     surla
    -0.49
     Biôgrafia
    -0.49
    POSITIVE LOGITS
     Audiodateien
    0.40
    errHandler
    0.35
    TRAILING
    0.34
    LEADING
    0.34
    SharedCtor
    0.32
    jeb
    0.32
    eb
    0.32
    
    0.31
    ógica
    0.31
     Vard
    0.31
    Activations Density 0.001%

    No Known Activations