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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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 - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 10397
    Prev
    Next
    INDEX
    Explanations

    references to "Sab" or variants thereof, suggesting it tracks mentions of a specific individual or entity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    starting with "Sab" or "Hab"

    np_acts-logits-general · gemini-2.0-flash

    Sabbats and names starting with Sab

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rrggbb
    -0.74
     resourceCulture
    -0.63
    ViewFeatures
    -0.62
    fromnode
    -0.61
    帖最后由
    -0.60
    +#+
    -0.60
    transQ
    -0.60
     بتاريخ
    -0.60
     pulmonar
    -0.59
     noqa
    -0.58
    POSITIVE LOGITS
     Sab
    0.76
    Sab
    0.73
    sab
    0.63
     SAB
    0.62
     sab
    0.60
    SAB
    0.59
    edoria
    0.48
     שוליים
    0.47
    displayquote
    0.46
     OFDb
    0.46
    Activations Density 0.157%

    No Known Activations