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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 694
    Prev
    Next
    INDEX
    Explanations

    references to geographical directions or regions, particularly related to the South and West

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Fridge
    -0.60
     Strafe
    -0.56
    box
    -0.56
    Box
    -0.55
    ListGroup
    -0.54
    cocho
    -0.54
     själ
    -0.54
    avice
    -0.53
    BOX
    -0.53
     Wikiseite
    -0.52
    POSITIVE LOGITS
    eastern
    0.71
    northern
    0.69
    ductory
    0.68
     Eastern
    0.66
    western
    0.66
     Northern
    0.65
    Northern
    0.63
    <bos>
    0.63
    Eastern
    0.63
     Southern
    0.63
    Activations Density 0.018%

    No Known Activations