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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 29774
    Prev
    Next
    INDEX
    Explanations

    the word "first" in various contexts indicating order or sequence

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

    first followed by an action or state

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    nalités
    -0.40
     getItemCount
    -0.40
    CELLANEOUS
    -0.38
    <?=
    -0.38
    rogels
    -0.38
    兩人
    -0.37
    aarrggbb
    -0.36
    orso
    -0.36
    évaluateur
    -0.36
    enumi
    -0.35
    POSITIVE LOGITS
     zuerst
    0.77
    abord
    0.73
     first
    0.72
    Primero
    0.72
     FIRST
    0.70
    pierw
    0.70
    先
    0.69
     First
    0.69
    First
    0.69
     先
    0.68
    Activations Density 0.035%

    No Known Activations