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 - 16k
    5. 1-GEMMASCOPE-MLP-16K
    6. 8773
    Prev
    Next
    INDEX
    Explanations

    items and descriptors related to food, specifically cheese and hot dogs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_mlp_out
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nahilalakip
    -0.96
    AndEndTag
    -0.96
    ьаж
    -0.91
    windowFixed
    -0.88
    脚注の使い方
    -0.81
    EndGlobalSection
    -0.79
    RetentionPolicy
    -0.77
     '{@
    -0.74
     للمعارف
    -0.73
     चीज़ों
    -0.72
    POSITIVE LOGITS
    '][]
    0.44
     savoury
    0.44
    ibat
    0.43
    úgó
    0.43
    |;
    0.43
     Tasty
    0.43
    uxxxx
    0.41
     afast
    0.41
    phant
    0.40
     Mérida
    0.39
    Activations Density 0.226%

    No Known Activations