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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14493
    Prev
    Next
    INDEX
    Explanations

    technical terms related to software testing and data structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Na
    -0.47
     Lamb
    -0.44
     NE
    -0.40
    Na
    -0.40
     Al
    -0.40
     NA
    -0.40
     Bird
    -0.39
    PASO
    -0.39
     Gla
    -0.38
     Advanced
    -0.38
    POSITIVE LOGITS
    TypedDataSet
    0.93
    principalTable
    0.59
     desmotivaciones
    0.55
    tagHelperRunner
    0.54
    Personensuche
    0.54
     Offisielt
    0.53
     sellado
    0.52
     coleção
    0.52
    Hentet
    0.52
     voorzien
    0.51
    Activations Density 1.625%

    No Known Activations