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. 10969
    Prev
    Next
    INDEX
    Explanations

    elements related to time-sensitive events or deadlines

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

    "today" and related temporal words

    np_acts-logits-general · gemini-2.0-flash

    today and tomorrow

    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
    Aktualisiert
    -0.49
    المشاركات
    -0.48
    nul
    -0.46
    Portail
    -0.46
    aurait
    -0.44
     matter
    -0.43
     ural
    -0.42
    likle
    -0.42
    neux
    -0.41
    endpush
    -0.41
    POSITIVE LOGITS
     today
    1.93
     tonight
    1.53
    today
    1.53
     tomorrow
    1.37
     TODAY
    1.36
     Today
    1.34
    Today
    1.30
     aujourd
    1.25
     heute
    1.25
    TODAY
    1.22
    Activations Density 0.171%

    No Known Activations