© 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. 33-GEMMASCOPE-RES-16K
    6. 11762
    Prev
    Next
    INDEX
    Explanations

    programming concepts, particularly those related to session management and configuration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    TypedDataSet
    -0.52
    setVerticalGroup
    -0.48
     ब्रेकडाउन
    -0.48
    səhifə
    -0.48
    RTLD
    -0.47
     AssemblyCompany
    -0.42
    يكب
    -0.41
     للاسماء
    -0.40
     AssemblyVersion
    -0.35
    IUrlHelper
    -0.35
    POSITIVE LOGITS
    tions
    1.12
    tion
    1.03
    tional
    0.96
    ions
    0.93
     tion
    0.92
     tions
    0.88
    TION
    0.86
    ion
    0.86
    IONS
    0.84
    ational
    0.81
    Activations Density 0.375%

    No Known Activations