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. 53173
    Prev
    Next
    INDEX
    Explanations

    citations or references to studies in scientific writing

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

    citations and gene regions

    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
     professionali
    -0.60
     oubli
    -0.52
     råd
    -0.48
    LayoutStyle
    -0.47
     morire
    -0.46
    siąż
    -0.45
     Bill
    -0.44
     étoit
    -0.43
    MigrationBuilder
    -0.43
    allclose
    -0.43
    POSITIVE LOGITS
    parsedMessage
    0.87
     Normdatei
    0.85
     OFDb
    0.76
    发表于
    0.72
     disambiguazione
    0.68
    ագրություններ
    0.66
     autorytatywna
    0.66
    nasse
    0.66
    enumii
    0.66
    ">//
    0.65
    Activations Density 0.090%

    No Known Activations