© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 1-GEMMASCOPE-RES-16K
    6. 3247
    Prev
    Next
    INDEX
    Explanations

    phrases indicating priority or emphasis, particularly in financial or decision-making contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    seamnă
    -0.57
    InjectAttribute
    -0.57
     resourceCulture
    -0.55
     unveils
    -0.52
    러나
    -0.52
     Música
    -0.52
    latego
    -0.52
     enamorado
    -0.52
     Wallflower
    -0.51
     Enkel
    -0.51
    POSITIVE LOGITS
    EL
    0.49
     btn
    0.47
     Priority
    0.46
     EL
    0.44
    btn
    0.41
    FirebaseUser
    0.41
     priority
    0.41
    findpost
    0.41
    AutoresizingMask
    0.40
    RTDA
    0.37
    Activations Density 0.155%

    No Known Activations