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. 10367
    Prev
    Next
    INDEX
    Explanations

    terms related to morality, faith, and justice

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

    Appears before "Offshore," or a virtue

    np_acts-logits-general · gemini-2.0-flash

    religious and positive concepts + 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
    •moral_agent

    No Comments

    Negative Logits
    StateException
    -0.72
    WindowConstants
    -0.64
    OGND
    -0.62
    masin
    -0.60
     betweenstory
    -0.59
    InSection
    -0.57
    notic
    -0.56
     kasarigan
    -0.54
    parsedMessage
    -0.54
    GTCX
    -0.53
    POSITIVE LOGITS
    fulness
    0.91
    fully
    0.85
     Majefty
    0.72
    FULLY
    0.69
    FACT
    0.69
    full
    0.69
     itſelf
    0.68
     Diſ
    0.66
     ſeveral
    0.65
    ously
    0.65
    Activations Density 0.386%

    No Known Activations