© 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. 34-GEMMASCOPE-RES-16K
    11. 3252
    Prev
    Next
    INDEX
    Explanations

    phrases related to analysis or evaluation of records and evidence

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.34.hook_resid_post
    Hook Layer
    34
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    имущества
    -0.46
    ViewFeatures
    -0.43
    LEncoder
    -0.43
    ulton
    -0.42
     SUS
    -0.42
    }{*}{
    -0.41
    ніципа
    -0.41
    hawat
    -0.41
    NameInMap
    -0.40
    sqlQuery
    -0.40
    POSITIVE LOGITS
     whole
    2.23
    whole
    2.11
    Whole
    1.92
     Whole
    1.73
     WHOLE
    1.72
     ganze
    1.26
     Ganzen
    1.22
    整体
    1.17
     geheel
    1.12
    整體
    1.10
    Activations Density 0.348%

    No Known Activations