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-2-2B
    4. MLP Out - 65k
    5. 18-GEMMASCOPE-MLP-65K
    6. 0
    Prev
    Next
    INDEX
    Explanations

    expressions of physical punishment and control in familial relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.18.hook_mlp_out
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BeginContext
    -0.85
    AddTagHelper
    -0.79
     المعيارى
    -0.79
     warrant
    -0.76
    TagMode
    -0.74
     warrants
    -0.73
     урна
    -0.73
    
    -0.69
    UnusedPrivate
    -0.69
     للمعارف
    -0.68
    POSITIVE LOGITS
    Beat
    0.62
     Beat
    0.57
     beat
    0.56
     beating
    0.53
    beat
    0.50
     sma
    0.50
     smack
    0.50
     whipping
    0.49
     beaten
    0.49
    beaten
    0.48
    Activations Density 0.105%

    No Known Activations