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. 7585
    Prev
    Next
    INDEX
    Explanations

    HTML and UI component structures within the code

    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
    TagMode
    -0.59
     betweenstory
    -0.54
    AndEndTag
    -0.49
     HasFactory
    -0.49
    balleur
    -0.46
    Personendaten
    -0.46
     ſch
    -0.44
     queſta
    -0.43
     exacta
    -0.43
    berdayakan
    -0.43
    POSITIVE LOGITS
     hype
    0.47
     Lottie
    0.46
    地下
    0.46
    lero
    0.45
     TCA
    0.45
    空
    0.45
     box
    0.44
     initComponents
    0.44
     tunnel
    0.43
     Chamber
    0.43
    Activations Density 0.092%

    No Known Activations