Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 10-GEMMASCOPE-RES-16K
    6. 16223
    Prev
    Next
    INDEX
    Explanations

    the beginning of a document or a new section

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aarrggbb
    -0.82
    RegressionTest
    -0.81
    }-*/;
    -0.70
    adaptiveStyles
    -0.70
    SOUNDBITE
    -0.69
     Мексичка
    -0.69
     ISNI
    -0.69
    kloped
    -0.68
     transfieras
    -0.67
     سكانية
    -0.66
    POSITIVE LOGITS
    enumi
    1.16
     gambe
    0.66
    enumerate
    0.61
    }{
    0.57
    masti
    0.54
    Ign
    0.53
    Rest
    0.53
    ınca
    0.53
    Repost
    0.52
    <h5>
    0.52
    Activations Density 0.003%

    No Known Activations