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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 10632
    Prev
    Next
    INDEX
    Explanations

    actions and states related to visual perception and observation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     FontWeight
    -0.78
     BoxFit
    -0.72
    rrggbb
    -0.66
    出版年
    -0.65
     Reverso
    -0.64
     OMIT
    -0.63
    shadowRadius
    -0.62
     Weich
    -0.62
    TagMode
    -0.62
     @"/
    -0.62
    POSITIVE LOGITS
     intently
    0.57
     blank
    0.57
     Richtung
    0.48
     horizon
    0.47
     direction
    0.47
    Πηγές
    0.47
    blank
    0.46
     faces
    0.45
     See
    0.43
     see
    0.42
    Activations Density 0.076%

    No Known Activations