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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 17-GEMMASCOPE-RES-65K
    6. 28551
    Prev
    Next
    INDEX
    Explanations

    function calls and event handling in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    клопе
    -0.94
    AndroidJUnit
    -0.92
    LookAnd
    -0.84
    ьаж
    -0.83
    oa̍t
    -0.81
    MigrationBuilder
    -0.81
    Datuak
    -0.78
    IUrlHelper
    -0.77
    WriteBarrier
    -0.76
    <_>
    -0.76
    POSITIVE LOGITS
    うちに
    0.46
    Object
    0.45
    触
    0.40
     event
    0.40
    うち
    0.40
    opo
    0.40
    物件
    0.38
     ev
    0.38
     events
    0.38
     Geschw
    0.38
    Activations Density 0.012%

    No Known Activations