© 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. 9951
    Prev
    Next
    INDEX
    Explanations

    references to environmental regulation and policing

    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
    0:0.08
    1:0.03
    2:0.04
    3:0.19
    4:0.16
    5:0.09
    6:0.07
    7:0.08
    8:0.03
    9:0.08
    10:0.04
    11:0.06
    Negative Logits
     Rutherford
    -3.50
     Michaels
    -3.00
     Ajax
    -2.90
    Wisconsin
    -2.85
     TBD
    -2.80
     Berks
    -2.78
     Shelby
    -2.74
    maxwell
    -2.73
    =#
    -2.71
    "]=>
    -2.60
    POSITIVE LOGITS
     Beijing
    5.95
     Taiwanese
    5.88
     Jiang
    5.63
     Shanghai
    5.60
     yuan
    5.50
     Xin
    5.47
    Hong
    5.46
     Chinese
    5.45
     Sina
    5.44
    Chinese
    5.44
    Activations Density 0.535%

    No Known Activations