Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 30632
    Prev
    Next
    INDEX
    Explanations

    references to the element iridium and its related properties or contexts

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

    words starting with "ir"

    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
    ReusableCell
    -0.61
     Pantal
    -0.54
     Bodega
    -0.51
    Samantha
    -0.48
    SaveChanges
    -0.48
    begingroup
    -0.48
    InputBorder
    -0.48
    Samp
    -0.47
    tagHelper
    -0.47
    ngdoc
    -0.46
    POSITIVE LOGITS
     Irvin
    0.85
     Ir
    0.82
     IR
    0.80
     Irwin
    0.76
     Irvine
    0.70
     Irma
    0.70
     Irons
    0.70
     ir
    0.69
     Irene
    0.68
    Ir
    0.68
    Activations Density 0.027%

    No Known Activations