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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 14546
    Prev
    Next
    INDEX
    Explanations

    programming constructs related to conditional statements and loops

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

    Assembly increment instructions

    np_acts-logits-general · gemini-2.0-flash

    swap operations

    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
    Sucesor
    -0.84
     ujednoznacz
    -0.67
    EndInit
    -0.66
     članak
    -0.65
     فريبيس
    -0.64
    Referanser
    -0.64
     ProtoMessage
    -0.63
    WriteBarrier
    -0.63
    ScopeManager
    -0.61
    Tracce
    -0.61
    POSITIVE LOGITS
    exchange
    0.80
     exchange
    0.78
    swap
    0.75
    Exchange
    0.72
     échange
    0.71
     swap
    0.70
    Swap
    0.68
     exchanges
    0.67
     intercambio
    0.64
     exchanging
    0.63
    Activations Density 0.312%

    No Known Activations