© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 105810
    Prev
    Next
    INDEX
    Explanations

    special characters and punctuation marks in mathematical and legal contexts

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

    Special punctuation and formatting symbols used in academic citations, mathematical equations, and legal references, particularly brackets, braces, and markup delimiters.

    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
     Administrativna
    -0.65
     Biôgrafia
    -0.64
    GEBURTSDATUM
    -0.64
     Italijanski
    -0.58
    /**
    -0.57
    WebElementEntity
    -0.56
     Италијани
    -0.56
    uxxxx
    -0.55
     Савезне
    -0.54
    elemField
    -0.50
    POSITIVE LOGITS
    ">//
    0.54
     Autorisations
    0.47
    httphttps
    0.47
    enumii
    0.47
    TagMode
    0.45
    enumi
    0.41
    出版年
    0.38
     שוליים
    0.38
    يميديا
    0.37
    ORPG
    0.37
    Activations Density 0.009%

    No Known Activations