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. 28499
    Prev
    Next
    INDEX
    Explanations

    phrases indicating actions of setting up or getting started

    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
     OMITBAD
    -0.63
    migrationBuilder
    -0.60
     hauts
    -0.58
     Rennen
    -0.57
    HasMaxLength
    -0.57
    Rüyada
    -0.56
     skolan
    -0.56
     mxArray
    -0.56
     "];
    -0.55
     discorso
    -0.53
    POSITIVE LOGITS
     وحد
    0.51
    ISupport
    0.48
     updated
    0.47
    getStyle
    0.47
    帳
    0.47
     fully
    0.46
     Sparrow
    0.46
    arted
    0.46
     current
    0.46
     Rah
    0.46
    Activations Density 0.112%

    No Known Activations