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. 12976
    Prev
    Next
    INDEX
    Explanations

    terms related to chemical reactions and synthesis processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     hran
    -0.54
    ঙ্গ
    -0.44
     player
    -0.43
    strconv
    -0.43
     quadril
    -0.42
     Interpret
    -0.42
     players
    -0.42
    zeczy
    -0.42
    isNaN
    -0.41
     חיצוניים
    -0.41
    POSITIVE LOGITS
    atalyst
    0.77
     catalyst
    0.75
    AsUp
    0.74
     catalysis
    0.74
    mybatisplus
    0.70
    TargetException
    0.69
     katal
    0.69
     Catalyst
    0.68
     فريبيس
    0.68
    省市镇
    0.68
    Activations Density 0.767%

    No Known Activations