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. 18-GEMMASCOPE-RES-16K
    6. 10459
    Prev
    Next
    INDEX
    Explanations

    phrases related to genetic pathways and their roles in diseases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     meist
    -0.61
     meistens
    -0.59
     Usually
    -0.56
    ⁡
    -0.56
     usually
    -0.56
    Биография
    -0.55
    usually
    -0.54
    だいたい
    -0.53
    Usually
    -0.52
     solito
    -0.52
    POSITIVE LOGITS
     future
    0.95
     potential
    0.85
     potentially
    0.85
    future
    0.81
    potential
    0.77
     künftig
    0.76
     promising
    0.76
    potentially
    0.74
     Potential
    0.73
     könnte
    0.71
    Activations Density 1.372%

    No Known Activations