© 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. 20-GEMMASCOPE-RES-16K
    11. 4463
    Prev
    Next
    INDEX
    Explanations

    key terms related to the names and origins of places and products.

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

    historical events and facts.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    instances where a name is changed or renamed. The neuron activates strongly for phrases indicating a change in name, such as "name was changed", "chose the name", "adopted the name", etc. It seems to be detecting situations where an entity, person, place, or thing is given

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    relating to name changes or naming

    np_acts-logits-general · gemini-2.0-flash

    company name changes

    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
    ItemLayout
    -0.53
    RenderAtEndOf
    -0.46
     estimés
    -0.41
    SetBool
    -0.40
     capaz
    -0.39
     тех
    -0.39
    complexContent
    -0.39
    CPtr
    -0.38
     VIDEOTAPE
    -0.38
    ElementException
    -0.38
    POSITIVE LOGITS
     naming
    1.37
     nickname
    1.34
     nicknames
    1.29
     renaming
    1.26
     rename
    1.18
     Naming
    1.18
     moniker
    1.16
    Naming
    1.13
    naming
    1.09
     surname
    1.09
    Activations Density 0.248%

    No Known Activations