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. 5698
    Prev
    Next
    INDEX
    Explanations

    technical specifications related to automotive components

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

    exhaust gas system

    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
    typeparam
    -0.44
    formik
    -0.44
    mobileqq
    -0.43
     wage
    -0.43
    Desk
    -0.42
     DESK
    -0.41
    IVATE
    -0.41
    blusa
    -0.40
    posas
    -0.40
     ferma
    -0.40
    POSITIVE LOGITS
     exhaust
    0.57
     Exhaust
    0.56
    Exhaust
    0.55
     muffler
    0.54
     tuyau
    0.54
    pipes
    0.53
     pipes
    0.52
    Ӕ
    0.50
    exhaust
    0.50
    principalTable
    0.49
    Activations Density 0.004%

    No Known Activations