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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 103267
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and relationships in equations

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

    Connecting phrases in scientific/mathematical text that describe how values are calculated, estimated, or related, often bridging between formal expressions.

    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
    RegistryLite
    -0.63
    GIVEREF
    -0.58
     (;;
    -0.54
     Scrip
    -0.53
     tripod
    -0.52
    ppuden
    -0.52
     CreateTagHelper
    -0.51
    gypti
    -0.50
    JNIEnv
    -0.49
    OGND
    -0.48
    POSITIVE LOGITS
    RenderAtEndOf
    0.56
     semula
    0.37
    InputTagHelper
    0.36
    AutoScale
    0.35
     Verhältnisse
    0.34
     bezpośred
    0.33
     heisst
    0.33
     daraus
    0.32
     Gerechtigkeit
    0.32
     rând
    0.32
    Activations Density 0.132%

    No Known Activations