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. 61602
    Prev
    Next
    INDEX
    Explanations

    terms related to evaluating and analyzing data relationships and operational commands

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

    commands, output, recognized, set

    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
     ſche
    -0.66
     ſch
    -0.59
    httphttps
    -0.56
    DoubleQuotes
    -0.54
    ArgsConstructor
    -0.54
    قایناق‌لار
    -0.53
     fevere
    -0.53
     שוליים
    -0.53
     viſ
    -0.52
     AssemblyTitle
    -0.51
    POSITIVE LOGITS
    insatz
    0.35
     дописавши
    0.33
    Somos
    0.32
     Sucher
    0.32
     Ehre
    0.31
     doigt
    0.31
     advertised
    0.30
    ByUser
    0.30
     honneur
    0.30
     Ahnung
    0.30
    Activations Density 0.261%

    No Known Activations