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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7552
    Prev
    Next
    INDEX
    Explanations

    references to knowledge and understanding, particularly in a contextual or technical setting

    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
     thái
    -0.36
    ellants
    -0.36
     whilst
    -0.35
    troppo
    -0.34
     द
    -0.34
    umerable
    -0.33
    packets
    -0.32
     efforts
    -0.32
    NotNil
    -0.32
    IVEREF
    -0.32
    POSITIVE LOGITS
     AssemblyCompany
    0.77
    sizeCache
    0.64
    IntoConstraints
    0.60
     ProtoMessage
    0.60
     intptr
    0.55
    parsedMessage
    0.52
    MessageTagHelper
    0.51
     considérons
    0.50
    сылкі
    0.49
    ValueStyle
    0.48
    Activations Density 0.419%

    No Known Activations