© 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. 2-GEMMASCOPE-RES-16K
    6. 8101
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and structures in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AssemblyCulture
    -0.72
     '\\;'
    -0.59
     AssemblyCompany
    -0.52
    bootstrapcdn
    -0.52
     medži
    -0.52
    RenderAtEndOf
    -0.51
    脚注の使い方
    -0.51
    áneo
    -0.50
    expandindo
    -0.49
    ientôt
    -0.48
    POSITIVE LOGITS
    CppMethod
    0.41
    MLLoader
    0.41
    }{*}{
    0.39
    IntoConstraints
    0.35
     hil
    0.35
    tuer
    0.34
    sap
    0.34
    swift
    0.33
    lavi
    0.33
     cult
    0.33
    Activations Density 0.244%

    No Known Activations