© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10606
    Prev
    Next
    INDEX
    Explanations

    references to magic and magical themes

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

    "magic" or its variants

    np_acts-logits-general · gemini-2.0-flash

    descriptions of magic

    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
    GEBURTS
    -0.70
     للاسماء
    -0.64
    KommentareTeilen
    -0.57
    存于互联网档案馆
    -0.56
     שוליים
    -0.54
    SequentialGroup
    -0.53
    <>("
    -0.53
     незавершена
    -0.53
    Issue
    -0.50
    estimable
    -0.50
    POSITIVE LOGITS
     magic
    1.36
     Magic
    1.21
    Magic
    1.12
     MAGIC
    1.10
    magic
    1.10
    MAGIC
    1.09
     magia
    1.05
     magical
    0.99
     Zauber
    0.96
     spells
    0.92
    Activations Density 0.109%

    No Known Activations