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. 10226
    Prev
    Next
    INDEX
    Explanations

    comparative expressions emphasizing superiority or inferiority

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

    Following "than" or "as"

    np_acts-logits-general · gemini-2.0-flash

    than or as comparisons

    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
     Aline
    -0.54
     Alger
    -0.52
     Florentine
    -0.50
     fuch
    -0.46
     uchiha
    -0.45
     isabel
    -0.44
     hana
    -0.44
     flori
    -0.43
     IDF
    -0.43
    ArrowToggle
    -0.43
    POSITIVE LOGITS
     ever
    1.14
     they
    1.09
     the
    0.91
     any
    0.91
     what
    0.91
    kyou
    0.90
     those
    0.87
     we
    0.86
     it
    0.85
     usual
    0.85
    Activations Density 0.150%

    No Known Activations