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. 24522
    Prev
    Next
    INDEX
    Explanations

    the Portuguese word "ao" and related variations, indicating a focus on specific terms or names in a language context

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

    ao identifiers and related syllables

    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
     lequel
    -0.53
    BindingResult
    -0.46
    Спољашње
    -0.46
     Méditerranée
    -0.45
     ویکی‌پدی
    -0.44
    subpackage
    -0.44
     périodes
    -0.43
     endroit
    -0.43
    – 
    -0.42
    ChildScrollView
    -0.40
    POSITIVE LOGITS
    ao
    1.06
    AO
    0.90
    Rao
    0.86
    Ao
    0.85
     Rao
    0.81
    hao
    0.79
     Ao
    0.77
    Pao
    0.76
    tao
    0.74
     AO
    0.74
    Activations Density 0.023%

    No Known Activations