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. 114508
    Prev
    Next
    INDEX
    Explanations

    references to military personnel and their actions

    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
     aufzu
    -0.34
     cheminée
    -0.32
     enfin
    -0.32
     merek
    -0.30
     pobre
    -0.30
     Staates
    -0.29
     enseignants
    -0.29
     nôtre
    -0.29
     dead
    -0.29
    founded
    -0.28
    POSITIVE LOGITS
     MainAxisSize
    0.77
    tanleria
    0.71
    endpush
    0.71
    IndentedString
    0.67
    Personendaten
    0.65
     noDo
    0.63
    MessageOf
    0.63
    SharedCtor
    0.61
    ngdoc
    0.61
     BoxFit
    0.60
    Activations Density 0.019%

    No Known Activations