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-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53027
    Prev
    Next
    INDEX
    Explanations

    technical terms related to electrical engineering and circuitry

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

    phase relationships

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    期刊论文
    -0.57
    nloa
    -0.56
     Neolithic
    -0.55
     Elizabethan
    -0.54
    ganu
    -0.54
     Thra
    -0.54
     δὲ
    -0.53
    RTRS
    -0.52
    republik
    -0.52
     SSE
    -0.52
    POSITIVE LOGITS
    WithIOException
    0.58
    ագրություններ
    0.57
     OFDb
    0.55
    ritsar
    0.53
     onCreateView
    0.50
    []):
    0.49
    ptonshire
    0.49
    IntoConstraints
    0.47
     ویکی‌پدیای
    0.47
     समीक्षक
    0.47
    Activations Density 0.071%

    No Known Activations