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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 4826
    Prev
    Next
    INDEX
    Explanations

    terms related to maintaining stability and preservation in systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Kim
    -0.59
     He
    -0.58
     rivi
    -0.58
    Ph
    -0.57
     fibras
    -0.57
     Bla
    -0.57
    Storyboard
    -0.57
     else
    -0.56
     UIKit
    -0.56
     pidana
    -0.56
    POSITIVE LOGITS
    INTAIN
    1.16
     upkeep
    1.10
     maintenance
    1.10
    intenance
    1.07
     onderhoud
    1.03
     maintained
    1.03
     Maintain
    1.03
     maint
    0.99
    Maintain
    0.98
     maintain
    0.98
    Activations Density 0.079%

    No Known Activations