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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 84
    Prev
    Next
    INDEX
    Explanations

    the term "mark" and its variations in the context of signaling or highlighting

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

    the word "mark" or variations of marking something.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    googleapis
    -0.52
    stdlib
    -0.49
    getSystem
    -0.48
    Util
    -0.48
    Yoo
    -0.46
     Yoo
    -0.44
     ích
    -0.44
     Gier
    -0.44
    Six
    -0.43
     gastroenter
    -0.43
    POSITIVE LOGITS
     mark
    1.07
     MARK
    1.05
     Mark
    0.98
    mark
    0.94
    MARK
    0.93
     Marks
    0.93
    Mark
    0.91
     marks
    0.90
     Marker
    0.88
    marks
    0.85
    Activations Density 0.136%

    No Known Activations