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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 13859
    Prev
    Next
    INDEX
    Explanations

    instances of family relationships and dynamics

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

    preceding or related to "arrive"

    np_acts-logits-general · gemini-2.0-flash

    arrival after entities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     disambiguazione
    -0.72
    Tikang
    -0.65
    knię
    -0.58
    esez
    -0.58
     архивлан
    -0.58
     snippetHide
    -0.57
    aronder
    -0.57
    ropia
    -0.54
    NameInMap
    -0.53
     Backman
    -0.52
    POSITIVE LOGITS
     arrive
    2.67
     arriving
    2.66
     arrival
    2.62
     arrived
    2.61
     arrives
    2.56
     Arrive
    2.24
     arrivals
    2.18
     Arrival
    2.18
    arrive
    2.17
    Arri
    2.15
    Activations Density 0.243%

    No Known Activations