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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 122671
    Prev
    Next
    INDEX
    Explanations

    structured data related to medical research and clinical trials

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

    Scientific or technical terminology appearing in academic or professional literature, including units of measurement, scientific methods, technical parameters, or specialized professional vocabulary.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     BytesLike
    -0.43
    Descripció
    -0.33
     connexe
    -0.32
    getVersion
    -0.30
     Unsc
    -0.28
    ritch
    -0.28
    okola
    -0.28
    wark
    -0.28
     lask
    -0.28
     quarters
    -0.28
    POSITIVE LOGITS
    EndContext
    0.69
     للمعارف
    0.63
     незавершена
    0.61
     ویکی‌آمباردا
    0.59
    findpost
    0.56
    +#+
    0.56
    IUrlHelper
    0.55
    DockStyle
    0.53
    IsContent
    0.51
     iprot
    0.51
    Activations Density 0.071%

    No Known Activations