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 - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 14396
    Prev
    Next
    INDEX
    Explanations

    technical terms related to software development, particularly those associated with binding and data models in programming contexts

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

    Period appearing at the end of a line

    np_acts-logits-general · gemini-2.0-flash

    xaml namespaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ber
    -0.52
    ros
    -0.49
    inst
    -0.49
    os
    -0.48
    mi
    -0.48
    ob
    -0.48
    se
    -0.47
    bu
    -0.47
    *)
    -0.46
    发表于
    -0.46
    POSITIVE LOGITS
     "..\..\..\
    0.88
     myſelf
    0.86
     cherchés
    0.80
     himſelf
    0.78
    NewUrlParser
    0.77
     itſelf
    0.75
     disambiguazione
    0.73
    ponses
    0.71
    CppMethod
    0.70
     Jefus
    0.70
    Activations Density 0.054%

    No Known Activations