© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 16759
    Prev
    Next
    INDEX
    Explanations

    references to Taiwan's political status and sentiments in relation to China

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.06
    1:0.02
    2:0.09
    3:0.05
    4:0.04
    5:0.06
    6:0.08
    7:0.05
    8:0.36
    9:0.09
    10:0.02
    11:0.02
    Negative Logits
     Arkham
    -3.68
     Diesel
    -3.63
    brance
    -3.60
    Arsenal
    -3.55
     mysql
    -3.54
     Satanic
    -3.53
    SELECT
    -3.44
     Salman
    -3.43
    Denver
    -3.37
     MPG
    -3.36
    POSITIVE LOGITS
     Taiwanese
    10.66
    Tai
    10.21
     Taiwan
    9.75
     Tai
    9.42
     Tao
    5.95
    pei
    5.64
     Huang
    5.57
     Tsu
    5.43
     Chung
    5.25
     Okinawa
    5.25
    Activations Density 0.043%

    No Known Activations