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. 29199
    Prev
    Next
    INDEX
    Explanations

    mentions of chemical compounds and their relevant contextual information

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

    company affiliations and locations

    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
    MessageTagHelper
    -0.32
     Autorizaciones
    -0.32
    
    -0.30
     and
    -0.28
     distributions
    -0.28
     Jerusal
    -0.27
    umoj
    -0.27
     Jérusalem
    -0.26
     useRouter
    -0.26
     rodeado
    -0.26
    POSITIVE LOGITS
     surla
    0.84
    存于互联网档案馆
    0.66
    ########.
    0.64
     ***!
    0.62
     مرئيه
    0.60
     autorytatywna
    0.57
    laken
    0.55
    ſelben
    0.55
     Waſſer
    0.54
     Мексичка
    0.54
    Activations Density 0.041%

    No Known Activations