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. 14309
    Prev
    Next
    INDEX
    Explanations

    discussions regarding criteria and physical experiences in the context of worship and health

    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
    ViewFeatures
    -0.54
    InitStruct
    -0.51
     BoxDecoration
    -0.50
    ibat
    -0.46
    idane
    -0.44
     spar
    -0.43
    mals
    -0.43
     jury
    -0.43
    aik
    -0.43
    +:+
    -0.42
    POSITIVE LOGITS
     physical
    0.65
    physical
    0.62
     körper
    0.59
     fisik
    0.58
     fysis
    0.56
     technical
    0.56
     technischen
    0.55
     physically
    0.53
    PHYSICAL
    0.52
    Physical
    0.52
    Activations Density 0.598%

    No Known Activations