© 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. 80552
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate relationships between variables or describe properties in mathematical and scientific contexts

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

    single, linear, convex, mean

    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
    Q
    -0.28
    ב
    -0.26
    ↵↵↵↵
    -0.25
    dotto
    -0.24
    eg
    -0.24
    verzeichnis
    -0.24
    __(
    -0.24
    Override
    -0.23
     takk
    -0.23
     fe
    -0.23
    POSITIVE LOGITS
    MessageOf
    0.86
    oredCriteria
    0.83
    XmlAccessorType
    0.73
    LookAnd
    0.69
    UrlResolution
    0.69
     <>",
    0.69
    AddTagHelper
    0.68
    ſicht
    0.66
     dieſes
    0.65
    ſchaft
    0.65
    Activations Density 0.095%

    No Known Activations