© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 909
    Prev
    Next
    INDEX
    Explanations

    references to modified processes or features in mathematical or scientific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    </i>
    -0.96
    </b>
    -0.85
    <i>
    -0.77
    क
    -0.72
    mtext
    -0.69
    iza
    -0.67
    <td>
    -0.67
    </td>
    -0.67
     cen
    -0.64
    아
    -0.64
    POSITIVE LOGITS
     مشين
    0.96
    SharedDtor
    0.95
     vuitton
    0.93
    المناصب
    0.91
     propOrder
    0.91
    tagHelperRunner
    0.89
     Waray
    0.89
     modding
    0.87
    Mods
    0.86
    firmasi
    0.86
    Activations Density 0.189%

    No Known Activations