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. 24
    Prev
    Next
    INDEX
    Explanations

    terms related to satisfaction, devotion, and possibly religious or spiritual concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    words or names related to "Sat" or "Satya".

    oai_token-act-pair · gpt-4oTriggered by @pigeonzow
    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
    ValueStyle
    -0.80
     ExecuteAsync
    -0.74
    
    -0.72
    SharedCtor
    -0.71
    COUVER
    -0.71
    
    -0.69
    хьтан
    -0.69
    WebVitals
    -0.68
    AxisAlignment
    -0.68
     InputDecoration
    -0.68
    POSITIVE LOGITS
     vicini
    0.46
     Concern
    0.44
     concern
    0.44
     TE
    0.44
    Concern
    0.44
     iprot
    0.43
     concerned
    0.43
     bug
    0.42
    applique
    0.42
     Bug
    0.41
    Activations Density 0.030%

    No Known Activations