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. 8043
    Prev
    Next
    INDEX
    Explanations

    biological and technical abbreviations or acronyms

    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
    cein
    -0.37
     supreme
    -0.36
    ulin
    -0.35
    ịch
    -0.34
     parting
    -0.33
    രിക്ക
    -0.32
     за
    -0.31
    <h2>
    -0.31
    visející
    -0.31
    udades
    -0.31
    POSITIVE LOGITS
    Personensuche
    0.62
     FP
    0.57
    richTextPanel
    0.57
    المشاركات
    0.56
     Cp
    0.56
    BufferException
    0.56
     myſelf
    0.54
    Fp
    0.52
     fp
    0.51
     rp
    0.51
    Activations Density 0.446%

    No Known Activations