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. 53009
    Prev
    Next
    INDEX
    Explanations

    references to scientific papers and their citation formats

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

    year and legal citation

    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
    DockStyle
    -0.56
    ్రామ
    -0.53
     ejus
    -0.53
    例句
    -0.51
     Espírito
    -0.48
    imov
    -0.48
    ATOS
    -0.47
     itſelf
    -0.47
    انجليز
    -0.46
     vechi
    -0.46
    POSITIVE LOGITS
    enumi
    0.82
    RenderAtEndOf
    0.79
     BorderRadius
    0.76
     ***!
    0.71
    expandindo
    0.67
    RTCK
    0.66
    存于互联网档案馆
    0.65
     CreateTagHelper
    0.64
     оригіналу
    0.63
    Portale
    0.63
    Activations Density 0.048%

    No Known Activations