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. 28557
    Prev
    Next
    INDEX
    Explanations

    instances of the special token used for beginning of segments or sections in a structured format

    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
    tagHelperRunner
    -1.01
     disambiguazione
    -0.98
    balleur
    -0.93
     Мексичка
    -0.81
    StoryboardSegue
    -0.81
    AccessorTable
    -0.74
    AsUp
    -0.73
     beginnetje
    -0.73
     springfox
    -0.72
     للمعارف
    -0.71
    POSITIVE LOGITS
     an
    0.72
     a
    0.69
     another
    0.54
     giúp
    0.51
     part
    0.49
     used
    0.49
     able
    0.43
     clés
    0.43
     beneficial
    0.43
     sebuah
    0.42
    Activations Density 0.189%

    No Known Activations