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. 85228
    Prev
    Next
    INDEX
    Explanations

    punctuation and formatting associated with citations or references

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

    location suffixes

    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
    transQ
    -0.86
    IntoConstraints
    -0.83
    évaluateur
    -0.78
    دانشنامهٔ
    -0.75
     transfieras
    -0.75
     CanadaChoose
    -0.73
    ंदीखरीदारी
    -0.70
     propOrder
    -0.69
     MainAxisSize
    -0.68
     незавершена
    -0.68
    POSITIVE LOGITS
    Web
    0.35
     tén
    0.34
     vocales
    0.30
     cuerdas
    0.29
     őket
    0.29
    bietet
    0.29
     kimse
    0.29
     Amérique
    0.29
    CC
    0.29
    mathop
    0.28
    Activations Density 0.038%

    No Known Activations