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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 43071
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "dirtiness" or "taintedness" in various contexts

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

    dirty blond, Dirty Three, dirty looks

    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
    Not in Any Lists

    No Comments

    Negative Logits
    IntoConstraints
    -0.66
     laſſen
    -0.63
    LayoutConstraint
    -0.62
    iibo
    -0.60
    rungsseite
    -0.60
    ſehen
    -0.60
     Pons
    -0.59
    volmente
    -0.58
    beelding
    -0.57
    liferay
    -0.57
    POSITIVE LOGITS
    Dirty
    1.13
    dirty
    1.13
     Dirty
    1.13
     dirty
    1.05
     dirt
    0.66
     Dirt
    0.66
     soiled
    0.65
    dirt
    0.65
     filthy
    0.65
     impure
    0.65
    Activations Density 0.005%

    No Known Activations