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. 12736
    Prev
    Next
    INDEX
    Explanations

    formatted output and structured data representations

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

    String formatting and related syntax

    np_acts-logits-general · gemini-2.0-flash

    format arguments

    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
     NSD
    -0.37
    RTDA
    -0.36
     BaseEntity
    -0.36
    chequer
    -0.35
    :✨
    -0.34
    perio
    -0.34
     initialState
    -0.33
     and
    -0.33
    elock
    -0.33
    bush
    -0.32
    POSITIVE LOGITS
    ValueStyle
    0.81
    ?}",
    0.81
     تضيفلها
    0.71
    enumii
    0.63
    !("{}",
    0.63
    下载附件
    0.61
    (@"%@",
    0.60
    NUMX
    0.59
     estekak
    0.58
     tabPage
    0.57
    Activations Density 0.065%

    No Known Activations