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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8670
    Prev
    Next
    INDEX
    Explanations

    specific scientific and legal terms related to trial evidence and methodologies

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    帖最后由
    -0.88
    sizeCache
    -0.69
    abestanden
    -0.68
    HasAnnotation
    -0.65
    .*")]
    -0.65
    endphp
    -0.65
    ″]
    -0.63
     disambiguazione
    -0.63
    AnchorTagHelper
    -0.62
    ]")]
    -0.61
    POSITIVE LOGITS
    CDCl
    0.35
     giudi
    0.33
    THO
    0.32
    UAGE
    0.31
    rijfs
    0.28
     descans
    0.28
     timmar
    0.28
    extré
    0.28
    zA
    0.27
    blij
    0.27
    Activations Density 1.139%

    No Known Activations