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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10840
    Prev
    Next
    INDEX
    Explanations

    punctuation and dialogue-related elements in text

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

    Attribution of quotes

    np_acts-logits-general · gemini-2.0-flash

    attribution verbs like says and said

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    JEL
    -0.43
    ătoare
    -0.40
    HasAnnotation
    -0.39
     Вікіпе
    -0.39
    googleapis
    -0.39
    ))/(
    -0.37
     vermek
    -0.36
    ))/
    -0.36
    bigsqcup
    -0.35
     kapal
    -0.35
    POSITIVE LOGITS
     says
    2.28
     said
    2.12
    says
    1.90
     Says
    1.77
    Says
    1.75
    said
    1.73
     say
    1.61
    Said
    1.54
     SAYS
    1.52
     Said
    1.37
    Activations Density 0.139%

    No Known Activations