© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 28032
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to spills and releases of substances, particularly in dangerous contexts

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

    items falling, thrown, or loose

    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
     industrielle
    -0.32
    '],
    -0.31
     lagi
    -0.30
     dispozici
    -0.30
    ۜ
    -0.29
     المقد
    -0.28
     Edison
    -0.27
    }]
    -0.27
     geïsole
    -0.26
    compile
    -0.26
    POSITIVE LOGITS
    verwijspagina
    0.66
    脚注の使い方
    0.65
    tvguidetime
    0.61
     otomatig
    0.60
     spillage
    0.59
    IsContent
    0.57
     scattering
    0.56
    Chham
    0.55
     dAtA
    0.55
    addCriterion
    0.55
    Activations Density 0.030%

    No Known Activations