© 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-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 42506
    Prev
    Next
    INDEX
    Explanations

    phrases indicating obligation or requirement

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    required to follow

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
    Negative Logits
     vecind
    -0.47
     idées
    -0.46
     logros
    -0.43
     pieles
    -0.43
     Idee
    -0.43
     descubrió
    -0.42
     democracia
    -0.42
     nubes
    -0.42
    joven
    -0.41
     otras
    -0.41
    POSITIVE LOGITS
     Must
    0.81
     must
    0.75
    Must
    0.75
    must
    0.73
     phải
    0.66
    MUST
    0.62
     MUST
    0.62
     Stiff
    0.60
     faudra
    0.59
     harus
    0.58
    Activations Density 0.022%

    No Known Activations