© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 2040
    Prev
    Next
    INDEX
    Explanations

    technical jargon related to computing and performance metrics

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

    Factors affecting performance or function

    np_acts-logits-general · gemini-2.0-flash

    performance, memory, nutrition, configuration

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     poffible
    -0.75
     myſelf
    -0.74
     reaſon
    -0.72
     fubject
    -0.70
     ſeveral
    -0.69
     chofe
    -0.67
     themſelves
    -0.66
     himſelf
    -0.66
     singola
    -0.65
     purpoſe
    -0.63
    POSITIVE LOGITS
    <bos>
    0.72
    bootstrapcdn
    0.69
    الإنجليزية
    0.62
    方面的
    0.59
    őre
    0.58
     issues
    0.58
    InSection
    0.58
    testens
    0.57
    epam
    0.56
     onAnimation
    0.55
    Activations Density 1.027%

    No Known Activations