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. 56534
    Prev
    Next
    INDEX
    Explanations

    requirements and processes related to evaluation and training in healthcare or engineering contexts

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

    recommendations to ensure

    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
    DockStyle
    -0.59
     BoxDecoration
    -0.44
    baomidou
    -0.43
     combineReducers
    -0.40
    saraba
    -0.39
    SaveChangesAsync
    -0.39
     JSONArray
    -0.38
     ovi
    -0.38
    ruptedException
    -0.38
    ReturnValue
    -0.37
    POSITIVE LOGITS
     adequate
    0.54
     preferably
    0.54
    preferably
    0.52
    要有
    0.50
     adecuados
    0.49
     Adequate
    0.49
     ensure
    0.49
     powinno
    0.48
     automatiques
    0.48
     adecuadas
    0.48
    Activations Density 0.024%

    No Known Activations