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. 13542
    Prev
    Next
    INDEX
    Explanations

    complex interactions and dynamics within narratives involving characters and themes

    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
    хьтан
    -0.52
     tartalmaz
    -0.49
    AndEndTag
    -0.48
     formado
    -0.44
     formada
    -0.44
     initComponents
    -0.44
     MaterialApp
    -0.43
     containing
    -0.42
    万能
    -0.42
    rekli
    -0.42
    POSITIVE LOGITS
     unheard
    1.30
     unprecedented
    1.15
     unmatched
    1.12
     unimaginable
    1.10
     unparalleled
    1.09
     rarely
    1.08
     unthinkable
    1.04
     seldom
    1.03
     unrival
    0.99
     uncommon
    0.99
    Activations Density 0.575%

    No Known Activations