© 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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1380
    Prev
    Next
    INDEX
    Explanations

    references to the pronoun "He."

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

    the pronoun "He"

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

    He followed by et

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    Примітки
    -0.46
     hang
    -0.43
     well
    -0.42
    конец
    -0.42
    nij
    -0.42
    ূ
    -0.42
    :✨
    -0.42
    ViewImports
    -0.41
     pluma
    -0.41
    [*
    -0.40
    POSITIVE LOGITS
    AndEndTag
    0.90
    copg
    0.83
     Meksiku
    0.77
    SourceChecksum
    0.71
     Exactos
    0.69
    (!__
    0.69
    otides
    0.68
     Италијани
    0.67
    ंदीखरीदारी
    0.66
    PMailer
    0.66
    Activations Density 0.021%

    No Known Activations