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. 8786
    Prev
    Next
    INDEX
    Explanations

    numerical values or measurements associated with components and assemblies

    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
    1
    -1.22
    4
    -0.89
    3
    -0.87
    2
    -0.84
    9
    -0.83
    7
    -0.82
    5
    -0.80
    6
    -0.77
    8
    -0.76
    0
    -0.63
    POSITIVE LOGITS
     utafitiHapana
    1.09
     nahilalakip
    1.05
    expandindo
    0.99
     للمعارف
    0.96
    parsedMessage
    0.95
     AssemblyCulture
    0.92
    <bos>
    0.91
     autorytatywna
    0.88
    -------
    0.88
    RenderAtEndOf
    0.88
    Activations Density 0.367%

    No Known Activations