© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 10373
    Prev
    Next
    INDEX
    Explanations

    code related to threading and synchronization mechanisms

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

    Code related to waiting or locks

    np_acts-logits-general · gemini-2.0-flash

    preventing and waiting

    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
    Parsed
    -0.46
    植
    -0.45
     Dedu
    -0.45
    Skipped
    -0.44
    лар
    -0.44
    SerializedSize
    -0.43
    Parse
    -0.43
    Buc
    -0.42
    pare
    -0.42
     beamed
    -0.41
    POSITIVE LOGITS
     waiting
    1.10
     wait
    1.03
     waits
    0.97
     waited
    0.96
     Waiting
    0.92
     GenerationType
    0.91
    WaitGroup
    0.91
     Wait
    0.89
     menunggu
    0.88
    waits
    0.87
    Activations Density 0.322%

    No Known Activations