Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 308
    Prev
    Next
    INDEX
    Explanations

    technical features and functionalities related to vehicle efficiency and operation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 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
    
    -0.41
     assembled
    -0.38
     kasarigan
    -0.38
     sheets
    -0.38
     pira
    -0.37
    ].(
    -0.36
     Sust
    -0.36
    ፩
    -0.35
     waxed
    -0.35
     контра
    -0.35
    POSITIVE LOGITS
     adaptive
    0.72
     automatic
    0.69
    adaptive
    0.69
     automática
    0.68
     automático
    0.67
    automatic
    0.66
    CloseOperation
    0.65
     automatically
    0.63
     Automatic
    0.62
    Automatic
    0.62
    Activations Density 1.184%

    No Known Activations