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. 80815
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "heart" as a metaphor for emotions and feelings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     Efq
    -0.54
     Blur
    -0.50
     للاسماء
    -0.50
    Література
    -0.49
     Lucien
    -0.47
     Gleaner
    -0.47
     Grooming
    -0.47
     Lesnar
    -0.46
     dentures
    -0.45
    basicConfig
    -0.45
    POSITIVE LOGITS
     heart
    1.05
    heart
    1.05
     hearts
    1.03
    Heart
    1.02
     corazón
    0.94
     Heart
    0.91
     HEART
    0.90
    corazón
    0.88
    hearts
    0.87
     cœurs
    0.86
    Activations Density 0.014%

    No Known Activations