© 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-IT
    7. Residual Stream - 131k
    9. 9-GEMMASCOPE-RES-131K
    11. 43592
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "pair" and references to tagging or identifying elements in a document

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

    C++ pair or HTML `src`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tradición
    -0.30
     vklju
    -0.29
     financiación
    -0.29
     toallas
    -0.29
     filosofía
    -0.28
     voces
    -0.27
     orgánico
    -0.27
     confirmación
    -0.27
     reputación
    -0.27
     adicionales
    -0.27
    POSITIVE LOGITS
    pair
    1.34
    quelize
    0.94
    featureID
    0.91
    PerformLayout
    0.90
    ransition
    0.89
    SequentialGroup
    0.86
     $_(
    0.79
     queſta
    0.78
    RegressionTest
    0.76
    WriteBarrier
    0.74
    Activations Density 0.004%

    No Known Activations