© 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. 4468
    Prev
    Next
    INDEX
    Explanations

    mentions of controversial events and figures, particularly related to satire and public reactions

    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
    0:0.25
    1:0.01
    2:0.14
    3:0.07
    4:0.03
    5:0.11
    6:0.03
    7:0.06
    8:0.16
    9:0.03
    10:0.04
    11:0.01
    Negative Logits
     Kush
    -3.00
     Kodi
    -2.93
     Marijuana
    -2.75
     Mariners
    -2.63
     Weed
    -2.58
    Tenn
    -2.52
     Starr
    -2.50
     Pine
    -2.49
    bilt
    -2.44
     Machina
    -2.43
    POSITIVE LOGITS
     Hebdo
    6.66
     cartoons
    4.20
     satirical
    3.89
     cartoon
    3.72
     satir
    3.48
     editorial
    3.41
     reprint
    3.41
     printing
    3.38
     Quran
    3.33
     caric
    3.27
    Activations Density 0.004%

    No Known Activations