Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 173
    Prev
    Next
    INDEX
    Explanations

    indications of surveillance or tracking through data analysis

    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
     EconPapers
    -0.60
    PMailer
    -0.53
    principalTable
    -0.52
    EndInit
    -0.50
    ///</
    -0.45
    __(/*!
    -0.44
    PerformLayout
    -0.44
     kaynağından
    -0.43
    awtextra
    -0.43
     nonUne
    -0.42
    POSITIVE LOGITS
     statistical
    0.57
     statistically
    0.54
     statistics
    0.54
     metrics
    0.50
     performance
    0.50
     stats
    0.49
     Statistical
    0.48
     usage
    0.46
     <=",
    0.44
    Statistical
    0.43
    Activations Density 0.700%

    No Known Activations