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. 13762
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal and ethical accountability in various contexts

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

    followed by "any" or "anything"

    np_acts-logits-general · gemini-2.0-flash

    do or change anything

    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
     autorytatywna
    -0.80
     ſeveral
    -0.70
     tambi
    -0.66
    сього
    -0.64
     alſo
    -0.63
    MessageTagHelper
    -0.62
     Wiktionnaire
    -0.61
     مشين
    -0.60
    Kaieteur
    -0.60
    AndGet
    -0.58
    POSITIVE LOGITS
     anything
    2.48
     any
    2.46
     anymore
    2.39
    anything
    2.14
     nor
    2.07
    any
    1.95
     anywhere
    1.94
     Anything
    1.88
    Anything
    1.85
     ANY
    1.85
    Activations Density 2.798%

    No Known Activations