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. 103009
    Prev
    Next
    INDEX
    Explanations

    terms related to various types of menus, quizzes, genres, and user roles in a coding or database context

    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
     Taktlose
    -0.66
     OMITBAD
    -0.66
    GTCX
    -0.63
    enterOuterAlt
    -0.63
    MigrationBuilder
    -0.63
     &___
    -0.59
    StoreMessageInfo
    -0.52
    EDEFAULT
    -0.52
     FetchType
    -0.51
     typelib
    -0.50
    POSITIVE LOGITS
    Personendaten
    0.43
     consist
    0.42
     consisting
    0.42
     actually
    0.40
    Shops
    0.40
    lette
    0.40
    HttpPut
    0.40
    สติ
    0.40
    他自己
    0.39
    InfoService
    0.39
    Activations Density 0.057%

    No Known Activations