© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 55007
    Prev
    Next
    INDEX
    Explanations

    references to YouTube and its related content

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

    YouTube

    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
     tarko
    -0.56
    makeConstraints
    -0.49
     aikaa
    -0.49
     NSCoder
    -0.46
     ricorda
    -0.46
     tärke
    -0.45
    Бахар
    -0.44
    BagConstraints
    -0.44
     vähän
    -0.43
     bewust
    -0.43
    POSITIVE LOGITS
     YouTube
    1.41
     youtube
    1.31
    YouTube
    1.30
     Youtube
    1.30
    youtube
    1.23
    Youtube
    1.22
     YOUTUBE
    1.13
    outube
    1.01
    YOUTUBE
    0.98
     YT
    0.82
    Activations Density 0.002%

    No Known Activations