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. 5-GEMMASCOPE-RES-16K
    6. 4304
    Prev
    Next
    INDEX
    Explanations

    instances of the word "to" with varying significance, indicating a focus on purpose or intention

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jména
    -1.00
     NDEBUG
    -0.73
    ation
    -0.70
    Namara
    -0.68
    jetbrains
    -0.65
     ilma
    -0.64
    گیر
    -0.64
     sähkö
    -0.64
    fisher
    -0.60
    StateChanged
    -0.59
    POSITIVE LOGITS
     to
    1.71
     TO
    1.44
     To
    1.24
    To
    1.22
    to
    1.18
     να
    1.11
     zu
    1.06
    TO
    1.06
    DataTo
    1.01
     לה
    0.97
    Activations Density 1.187%

    No Known Activations