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. 2956
    Prev
    Next
    INDEX
    Explanations

    statements indicating the versatility and functionality of technology

    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
    Надо
    -1.10
     Надо
    -1.07
     daß
    -0.98
    Herzliche
    -0.98
     Daß
    -0.97
     freilich
    -0.96
     היתה
    -0.89
    けっこう
    -0.89
     guère
    -0.89
     надо
    -0.88
    POSITIVE LOGITS
    Additionally
    0.92
     Additionally
    0.87
    Lastly
    0.82
    除此之外
    0.81
    Whilst
    0.80
    Furthermore
    0.79
     Lastly
    0.78
     jedoch
    0.75
    Upon
    0.75
    Throughout
    0.75
    Activations Density 0.176%

    No Known Activations