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. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 2951
    Prev
    Next
    INDEX
    Explanations

    content related to time-sensitive events and their resolutions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    __':
    -0.69
    __":
    -0.65
     تانيه
    -0.62
    cshtml
    -0.60
    Sortie
    -0.56
     StatefulWidget
    -0.56
    istoitu
    -0.54
     StatelessWidget
    -0.53
    enumi
    -0.53
     BoxDecoration
    -0.51
    POSITIVE LOGITS
    PHI
    0.60
     Phil
    0.54
     Calam
    0.54
    pushFollow
    0.53
     quia
    0.53
    phils
    0.52
     resourceCulture
    0.52
     occa
    0.50
     Aurora
    0.50
     philipp
    0.49
    Activations Density 0.043%

    No Known Activations