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. 14534
    Prev
    Next
    INDEX
    Explanations

    syntax and structure in code snippets

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

    Brackets and colons

    np_acts-logits-general · gemini-2.0-flash

    code structure and delimiters

    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
     quæ
    -0.50
     bì
    -0.49
     circon
    -0.48
     couver
    -0.48
    figura
    -0.48
     Included
    -0.47
     innova
    -0.46
     Covered
    -0.46
     allora
    -0.45
     couvert
    -0.45
    POSITIVE LOGITS
    abestanden
    0.86
    WriteBarrier
    0.81
     Signalez
    0.73
    ymce
    0.72
    verwijspagina
    0.71
    adaptiveStyles
    0.68
    __(/*!
    0.66
    sizeCache
    0.65
     protoimpl
    0.64
     (£
    0.63
    Activations Density 0.303%

    No Known Activations