© 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. 21-GEMMASCOPE-RES-16K
    6. 10812
    Prev
    Next
    INDEX
    Explanations

    references to collaborative research and partnerships

    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
     oby
    -0.32
    CNC
    -0.31
     DCM
    -0.30
    TOUCH
    -0.30
     doigt
    -0.30
     Zor
    -0.29
     ahorro
    -0.29
     food
    -0.29
     ord
    -0.28
     dru
    -0.28
    POSITIVE LOGITS
    实验室
    0.70
    RectangleBorder
    0.67
     laboratory
    0.64
    
    0.64
     scientific
    0.63
     scientist
    0.63
    Биография
    0.61
     laboratories
    0.61
     collaborators
    0.61
     researcher
    0.60
    Activations Density 1.426%

    No Known Activations