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. 8806
    Prev
    Next
    INDEX
    Explanations

    names of artists and their connections to various works or roles they have played

    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
    labus
    -0.91
    ^(@)
    -0.89
    mybatisplus
    -0.85
    òria
    -0.85
    contentLoaded
    -0.84
    WithIOException
    -0.84
     BorderRadius
    -0.84
    sphase
    -0.81
    resaid
    -0.81
     Walkover
    -0.79
    POSITIVE LOGITS
    ers
    0.37
    Bode
    0.35
    PHPExcel
    0.35
    -
    0.34
    od
    0.34
    Kier
    0.34
    us
    0.34
     hierogly
    0.33
     streets
    0.33
    men
    0.32
    Activations Density 8.782%

    No Known Activations