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. 4820
    Prev
    Next
    INDEX
    Explanations

    references to the word "after"

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

    "After" as a temporal marker

    np_acts-logits-general · gemini-2.0-flash

    after all

    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
    Hochspringen
    -0.77
     ་་
    -0.75
     dAtA
    -0.70
    PreferredItem
    -0.69
    cokinetics
    -0.69
    kheim
    -0.68
     intStringLen
    -0.68
     विश्वसनीयता
    -0.67
     Jefus
    -0.66
    IUrlHelper
    -0.66
    POSITIVE LOGITS
    words
    0.82
    thought
    0.78
    thoughts
    0.78
    word
    0.77
    glow
    0.70
    effects
    0.70
     After
    0.69
    care
    0.69
    taste
    0.68
    wards
    0.67
    Activations Density 0.157%

    No Known Activations