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-2-9B-IT
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 6532
    Prev
    Next
    INDEX
    Explanations

    terms related to nuclear energy and its implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    following "nuclear"

    np_acts-logits-general · gemini-2.0-flash

    nuclear science and applications

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AspNetCore
    -0.46
     optar
    -0.41
     gap
    -0.40
     lifestyle
    -0.40
    相反
    -0.40
     prior
    -0.40
     Gap
    -0.39
    itiert
    -0.39
    prior
    -0.38
     Mercier
    -0.37
    POSITIVE LOGITS
     nuclear
    1.95
    Nuclear
    1.87
     Nuclear
    1.84
    nuclear
    1.77
     NUCLEAR
    1.70
     nucléaire
    1.59
     nukes
    1.32
     nuke
    1.30
     Nucl
    1.23
     ядер
    1.23
    Activations Density 0.386%

    No Known Activations