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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114659
    Prev
    Next
    INDEX
    Explanations

    transitional phrases and conjunctions that connect thoughts and ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     Мексичка
    -0.87
    KommentareTeilen
    -0.83
    transQ
    -0.81
    onViewCreated
    -0.79
    ScopeManager
    -0.77
    IUrlHelper
    -0.77
    RegressionTest
    -0.76
     betweenstory
    -0.75
     geſch
    -0.73
    ikusbot
    -0.73
    POSITIVE LOGITS
     darse
    0.30
    ,
    0.28
    jalá
    0.27
     suaminya
    0.27
     initially
    0.26
     salido
    0.26
     jelasnya
    0.26
     Initially
    0.26
    tualmente
    0.26
     Ähn
    0.25
    Activations Density 0.027%

    No Known Activations