© 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. 10-ATT-KK
    11. 16462
    Prev
    Next
    INDEX
    Explanations

    references to unjustified violence or killings, particularly in a political context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.32
    1:0.05
    2:0.04
    3:0.06
    4:0.05
    5:0.04
    6:0.09
    7:0.08
    8:0.02
    9:0.06
    10:0.06
    11:0.07
    Negative Logits
     Budapest
    -4.34
     Slovakia
    -3.90
     Hungary
    -3.79
     Hitler
    -3.74
     Berlin
    -3.72
     �
    -3.69
     Cologne
    -3.60
     Hungarian
    -3.59
     Nazi
    -3.53
     Jericho
    -3.47
    POSITIVE LOGITS
     Aw
    5.08
     Yemeni
    4.71
    Aw
    4.59
     Imran
    4.45
     AQ
    4.14
     drone
    4.08
     Yemen
    3.89
     Pakistani
    3.72
     Drone
    3.67
     Brennan
    3.65
    Activations Density 0.003%

    No Known Activations