© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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 - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 6843
    Prev
    Next
    INDEX
    Explanations

    questions and decision-making processes in medical and treatment contexts

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

    Comes before qualifying words or phrases

    np_acts-logits-general · gemini-2.0-flash

    safely confident

    np_acts-logits-general · gemini-2.5-flash-lite

    language about timing and rules—durations/intervals and their extension or scheduling, plus conditional permissions/exclusions (e.g., when something is allowed, unnecessary, or ineligible).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    Not in Any Lists

    No Comments

    Negative Logits
    contentLoaded
    -0.55
     llanto
    -0.32
    참고
    -0.32
     disque
    -0.29
     roue
    -0.28
    popularity
    -0.28
     chia
    -0.28
    
    -0.28
    :"-"`
    -0.27
     unprofitable
    -0.27
    POSITIVE LOGITS
     safely
    0.59
    tagHelperRunner
    0.58
    parsedMessage
    0.53
    ########.
    0.53
     permitting
    0.53
     confidently
    0.52
    許
    0.52
    exitRule
    0.51
    Confidence
    0.50
    confident
    0.50
    Activations Density 0.541%

    No Known Activations