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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 102966
    Prev
    Next
    INDEX
    Explanations

    transitional phrases and numerical values related to conditions or calculations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    IsMutable
    -0.52
     وتسجيلات
    -0.48
     propOrder
    -0.47
     فريبيس
    -0.47
     createSlice
    -0.43
    MessageTagHelper
    -0.41
    (!__
    -0.41
    synthesize
    -0.40
     resourceCulture
    -0.40
    ----</
    -0.40
    POSITIVE LOGITS
     noticia
    0.42
     oscuridad
    0.42
     actualité
    0.41
     Pingback
    0.41
    勤
    0.38
     trabajos
    0.38
    umumkan
    0.38
     reacties
    0.38
     ingeniería
    0.37
    resultados
    0.37
    Activations Density 0.239%

    No Known Activations