© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 56375
    Prev
    Next
    INDEX
    Explanations

    initial tokens that signal the beginning of a text or section

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BASELINE
    -0.64
     INTERESAR
    -0.60
     dụng
    -0.58
     Stoll
    -0.49
    WriteString
    -0.48
    ⓘ
    -0.48
    exitRule
    -0.47
    sizeCache
    -0.47
    inguished
    -0.46
     initComponents
    -0.46
    POSITIVE LOGITS
     efforts
    0.72
    efforts
    0.66
     effort
    0.66
     endeavours
    0.58
     endeavour
    0.57
     esfuerzos
    0.55
     strain
    0.53
     esforço
    0.53
     Hec
    0.53
    DrawerToggle
    0.52
    Activations Density 0.045%

    No Known Activations