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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 27693
    Prev
    Next
    INDEX
    Explanations

    sections of text that reference specific dates or date formats

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

    dates and times

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    magnetism
    -0.54
     Magnetism
    -0.52
    wachsene
    -0.51
    wachsenen
    -0.50
    ğraf
    -0.49
    Magne
    -0.47
     ménage
    -0.46
    variates
    -0.46
    hibli
    -0.45
    úgó
    -0.45
    POSITIVE LOGITS
     الاطلاع
    0.77
    GEBURTSDATUM
    0.74
     وتسجيلات
    0.67
    MemoryWarning
    0.66
     يتيمه
    0.65
    fjspx
    0.65
     MaterialApp
    0.64
    Datuak
    0.63
     confronti
    0.62
     Portail
    0.60
    Activations Density 0.018%

    No Known Activations