© 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
    7. Residual Stream - 131k
    9. 24-GEMMASCOPE-RES-131K
    11. 707
    Prev
    Next
    INDEX
    Explanations

    technical or programming-related terms and syntax elements

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

    struct field definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    richTextPanel
    -0.43
     pageContext
    -0.39
    getWriter
    -0.39
    makeText
    -0.38
     elä
    -0.38
    发表于
    -0.37
     yaşad
    -0.37
    reactstrap
    -0.37
     OMIT
    -0.36
     تكبرها
    -0.36
    POSITIVE LOGITS
     propOrder
    0.60
    ConstraintMaker
    0.44
    ighborhood
    0.43
    optional
    0.43
    GEBURTSDATUM
    0.43
     uLocal
    0.42
    takan
    0.41
     optional
    0.41
    domada
    0.41
    coltà
    0.40
    Activations Density 0.060%

    No Known Activations