© 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. 31-GEMMASCOPE-RES-131K
    11. 81126
    Prev
    Next
    INDEX
    Explanations

    clauses that indicate legal requirements and obligations

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

    legal and regulatory compliance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    InjectMocks
    -0.55
    MLLoader
    -0.53
    évaluateur
    -0.50
     defaultstate
    -0.48
    addContainerGap
    -0.48
     ruddy
    -0.46
     poetical
    -0.46
     bushels
    -0.46
     curiously
    -0.46
     vectorielles
    -0.45
    POSITIVE LOGITS
     (§
    0.55
     pursuant
    0.54
     legally
    0.45
     statutory
    0.44
     agreed
    0.44
     obligated
    0.43
     obliged
    0.42
     notified
    0.42
     enforceable
    0.40
     lawfully
    0.40
    Activations Density 0.016%

    No Known Activations