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. 15-GEMMASCOPE-RES-16K
    6. 2351
    Prev
    Next
    INDEX
    Explanations

    scientific terms and abbreviations related to various biological and chemical 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
     ་་
    -1.31
     itſelf
    -1.24
     ―――――
    -1.20
    >\<^
    -1.03
    )");
    
    -1.00
     iſt
    -0.99
     faſt
    -0.95
     ſind
    -0.94
    ſelf
    -0.94
    Tikang
    -0.94
    POSITIVE LOGITS
    p
    0.65
    MS
    0.62
     område
    0.62
     vendus
    0.61
    SP
    0.59
     membunuh
    0.59
     M
    0.59
     soeur
    0.59
    w
    0.58
    f
    0.57
    Activations Density 1.246%

    No Known Activations