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. 20-GEMMASCOPE-RES-16K
    6. 347
    Prev
    Next
    INDEX
    Explanations

    phrases and concepts related to accountability and compliance

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

    Prepositions such as "to," "of," "about," "with," "on," and "for"

    np_acts-logits-general · gemini-2.0-flash

    prepositions followed by it

    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
    ؤلاء
    -0.64
     NSCoder
    -0.60
    StructEnd
    -0.60
    Distribuzione
    -0.55
    StoryboardSegue
    -0.54
     protoimpl
    -0.54
    зулта
    -0.53
    NOPQRST
    -0.53
    qrstuvwxyz
    -0.53
     تضيفلها
    -0.51
    POSITIVE LOGITS
     it
    3.10
     them
    2.27
     ervan
    2.00
    它
    1.89
     thereof
    1.88
     оно
    1.77
     esso
    1.77
     nó
    1.72
     آن
    1.68
     dalamnya
    1.68
    Activations Density 2.658%

    No Known Activations