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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 12708
    Prev
    Next
    INDEX
    Explanations

    specific symbols and formatting in scientific notation or identifiers

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

    Superscripted numbers

    np_acts-logits-general · gemini-2.0-flash

    mathematical notation and symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    .*")]
    -0.66
    ConstraintMaker
    -0.56
    +:+
    -0.53
    transQ
    -0.53
    fromnode
    -0.46
     betweenstory
    -0.45
    ItemBackground
    -0.45
    richTextPanel
    -0.39
     Driscoll
    -0.39
     connections
    -0.38
    POSITIVE LOGITS
     autorytatywna
    0.77
     surla
    0.57
     Италијани
    0.56
     كومونز
    0.54
    tagext
    0.54
     ब्रेकडाउन
    0.52
    ⓧ
    0.52
    хьтан
    0.51
    FormTagHelper
    0.50
     Inſ
    0.49
    Activations Density 0.702%

    No Known Activations