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. 6-GEMMASCOPE-MLP-16K
    6. 163
    Prev
    Next
    INDEX
    Explanations

    phrases associated with regulatory or forward-looking statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.6.hook_mlp_out
    Hook Layer
    6
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     HasFactory
    -0.64
     nahilalakip
    -0.55
     cavalli
    -0.52
    եղ
    -0.52
    oneof
    -0.48
     tearDown
    -0.47
    reat
    -0.47
     vực
    -0.46
     hang
    -0.46
     capital
    -0.45
    POSITIVE LOGITS
     getRule
    0.69
    ruptedException
    0.64
     bezeichneter
    0.63
     MainAxisSize
    0.63
    σμα
    0.62
    Cyfeiriadau
    0.61
    PositiveButton
    0.60
     CreateTagHelper
    0.60
    issime
    0.60
    tzmann
    0.59
    Activations Density 0.093%

    No Known Activations