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. 7682
    Prev
    Next
    INDEX
    Explanations

    conjunctions and logical connectors in sentences

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

    Relative pronouns ("which," "that," "who," "whom," "whose")

    np_acts-logits-general · gemini-2.0-flash

    relative pronouns and conjunctions

    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
     EconPapers
    -0.60
    Extinguishing
    -0.60
    assertNull
    -0.52
    ndar
    -0.50
    domada
    -0.49
    fycat
    -0.48
     there
    -0.48
    ::*;
    -0.47
    儡
    -0.47
    ']").
    -0.47
    POSITIVE LOGITS
     which
    2.40
     whose
    2.18
    which
    1.91
    whose
    1.90
     who
    1.89
     whom
    1.74
     WHICH
    1.62
     Whose
    1.62
     cuyas
    1.57
     cuyo
    1.57
    Activations Density 0.292%

    No Known Activations