Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 398
    Prev
    Next
    INDEX
    Explanations

    debugging messages related to errors in a system or application

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

    log messages related to debugging

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

    DEBUG statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     resourceCulture
    -0.90
     propOrder
    -0.87
    WindowConstants
    -0.86
    XmlEnum
    -0.81
    ChromeDriver
    -0.80
    énario
    -0.79
    RegistryLite
    -0.79
    脚注の使い方
    -0.75
     AttributeSet
    -0.75
    وانید
    -0.74
    POSITIVE LOGITS
    DEBUG
    0.59
     DEBUG
    0.54
    displaymath
    0.51
    ship
    0.48
    readLine
    0.47
    bitset
    0.46
    ing
    0.46
    koku
    0.45
    hower
    0.45
     suspect
    0.45
    Activations Density 0.010%

    No Known Activations