© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 21-GEMMASCOPE-RES-16K
    11. 10498
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and notations, particularly those involving inequalities and definitions related to sets or subsets

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    principalTable
    -0.33
     byen
    -0.32
     следует
    -0.31
    I
    -0.31
     devriez
    -0.31
     polegadas
    -0.30
     bertanggung
    -0.30
     öld
    -0.30
     ljudi
    -0.29
     zufolge
    -0.29
    POSITIVE LOGITS
     kaarangay
    0.78
     ErrIntOverflow
    0.70
     nakalista
    0.68
     חיצוניים
    0.67
    astify
    0.63
    Diweddarwch
    0.63
    хьтан
    0.62
    Bibliograf
    0.62
    QMetaType
    0.62
     ویکی‌پدی
    0.62
    Activations Density 1.244%

    No Known Activations