Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24-GEMMASCOPE-RES-16K
    6. 4446
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the letter "R"

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

    "R" followed by punctuation or other characters

    np_acts-logits-general · gemini-2.0-flash

    standalone R letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    heça
    -0.66
    kloped
    -0.63
    findpost
    -0.62
    ultado
    -0.53
    UnitTesting
    -0.52
    nyttet
    -0.51
    Varint
    -0.50
     lenker
    -0.50
    kuuta
    -0.50
     typelib
    -0.49
    POSITIVE LOGITS
     سكانية
    0.57
    oosevelt
    0.52
    osh
    0.52
    aptors
    0.51
    itter
    0.51
    ector
    0.51
     beginnetje
    0.51
    mises
    0.50
    ússia
    0.50
     Потому
    0.49
    Activations Density 0.168%

    No Known Activations