© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 83818
    Prev
    Next
    INDEX
    Explanations

    references to time periods, specifically focusing on expressions related to quantity and duration, such as "few" and "past."

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

    quantified time periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     betweenstory
    -0.76
     later
    -0.46
     transférez
    -0.43
    その後
    -0.42
     später
    -0.42
     Later
    -0.41
     âu
    -0.39
    後來
    -0.38
     lenker
    -0.37
     späteren
    -0.37
    POSITIVE LOGITS
     weekends
    0.69
     Wednesdays
    0.66
     month
    0.64
     weeks
    0.64
     Thursdays
    0.63
     year
    0.63
    Thursday
    0.62
     Fridays
    0.61
     Friday
    0.60
     week
    0.60
    Activations Density 0.017%

    No Known Activations