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. 2218
    Prev
    Next
    INDEX
    Explanations

    references to political events and figures

    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
     AssemblyProduct
    -0.56
    rungsseite
    -0.56
    optarg
    -0.54
    libft
    -0.54
    bkz
    -0.53
     noDo
    -0.51
    Hochspringen
    -0.51
     kapturem
    -0.50
     ويكيميديا
    -0.49
     femininos
    -0.49
    POSITIVE LOGITS
    Second
    0.41
     Second
    0.40
     بيها
    0.38
    MENAFN
    0.37
    Outside
    0.37
     Outside
    0.36
     Meanwhile
    0.36
     сно
    0.35
     Past
    0.35
    external
    0.35
    Activations Density 0.332%

    No Known Activations