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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 1260
    Prev
    Next
    INDEX
    Explanations

    text that connects to specific actions or items in a computational or scientific context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mogat
    -0.53
     Zellen
    -0.53
    UrlResolution
    -0.53
     schimb
    -0.53
     snippetHide
    -0.53
    وردار
    -0.51
     neige
    -0.51
    optionalTypeArgs
    -0.50
    marle
    -0.50
     izd
    -0.48
    POSITIVE LOGITS
    awtextra
    0.71
     للاسماء
    0.61
     ffilmiau
    0.59
    freopen
    0.57
    hyrchwyd
    0.57
    ScopeManager
    0.57
    AndroidJUnit
    0.57
    Parcelize
    0.56
    เอา
    0.55
    }{*}{}
    0.55
    Activations Density 0.467%

    No Known Activations