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. 122660
    Prev
    Next
    INDEX
    Explanations

    quantitative data related to health and disease in experimental settings

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

    Scientific notation in mathematical formulas and academic papers, particularly gender symbols (♂, ♀) identifying male or female subjects in experimental research studies, citation markers (i.e. [@BXX]), and mathematical equation formatting elements like commas, brackets, and text styling commands.

    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
     GenerationType
    -0.83
    uxxxx
    -0.61
    
    -0.56
    MigrationBuilder
    -0.56
     ویکی‌پدیای
    -0.52
    expandindo
    -0.52
    copg
    -0.49
    iblichen
    -0.49
    QMetaType
    -0.47
    GenerationType
    -0.46
    POSITIVE LOGITS
    ℠
    0.45
     nominally
    0.44
    ):=\
    0.44
     («
    0.42
     („
    0.42
    '&:
    0.42
    这意味着
    0.41
    ‽
    0.41
    クルマ
    0.40
    */),
    0.39
    Activations Density 0.440%

    No Known Activations