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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 6014
    Prev
    Next
    INDEX
    Explanations

    terms related to elevated levels and measurements in medical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    elevated, high expression, overexpression

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     transform
    -0.50
     transitions
    -0.46
     elegantly
    -0.46
     transforms
    -0.44
     rever
    -0.44
    artamentos
    -0.44
     pud
    -0.43
     settling
    -0.42
     reinvent
    -0.42
    Pfe
    -0.42
    POSITIVE LOGITS
    posedge
    0.69
     الرياضيه
    0.55
    Tikang
    0.52
    Excessive
    0.50
     anormal
    0.49
    addCriterion
    0.47
     exceso
    0.46
     excesso
    0.43
     abundancia
    0.43
     excès
    0.42
    Activations Density 0.207%

    No Known Activations