Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 369
    Prev
    Next
    INDEX
    Explanations

    indicative phrases or expressions related to uncertainties or possibilities

    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
    océan
    -0.40
     propia
    -0.36
    feitura
    -0.35
    經驗
    -0.35
     propio
    -0.34
     Doch
    -0.34
    beginTransaction
    -0.34
     понравился
    -0.32
    astify
    -0.32
     ar
    -0.31
    POSITIVE LOGITS
    UserScript
    0.73
    uxxxx
    0.71
    protoimpl
    0.71
     Italijani
    0.64
    httphttps
    0.60
     ſche
    0.58
    autoreleasepool
    0.58
     Anſ
    0.57
     Biôgrafia
    0.56
    astéroïdes
    0.55
    Activations Density 1.353%

    No Known Activations