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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 36-GEMMASCOPE-RES-16K
    6. 14685
    Prev
    Next
    INDEX
    Explanations

    specific programming constructs and variable definitions in code snippets

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     judge
    -1.38
     judges
    -1.26
     Judge
    -1.20
    judge
    -1.20
     JUDGE
    -1.13
    Judge
    -1.13
     judged
    -1.11
     judgment
    -1.11
    jud
    -1.10
     JUD
    -1.08
    POSITIVE LOGITS
    Tembelea
    0.47
    ⓧ
    0.43
     grà
    0.38
    
    0.38
    Xna
    0.37
     Monfieur
    0.36
    Evan
    0.35
    elesaikan
    0.34
    postsleuth
    0.34
    ",$
    0.34
    Activations Density 0.265%

    No Known Activations