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. 8852
    Prev
    Next
    INDEX
    Explanations

    emphasizing words that denote intensity or degree

    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
    MigrationBuilder
    -1.06
    ########.
    -1.04
     CURIAM
    -0.94
    Portail
    -0.93
     Exacts
    -0.91
    enumi
    -0.91
    __':
    
    -0.91
     للاسماء
    -0.88
    UserScript
    -0.86
    IntoConstraints
    -0.84
    POSITIVE LOGITS
     often
    0.45
     most
    0.44
     able
    0.39
     not
    0.38
     usually
    0.36
     ellers
    0.36
     fejl
    0.35
    often
    0.34
     opgenomen
    0.34
     part
    0.33
    Activations Density 0.078%

    No Known Activations