© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 10-ATT-KK
    11. 11959
    Prev
    Next
    INDEX
    Explanations

    statements related to legal charges or criminal activities

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.05
    1:0.05
    2:0.06
    3:0.06
    4:0.04
    5:0.18
    6:0.11
    7:0.07
    8:0.07
    9:0.14
    10:0.07
    11:0.05
    Negative Logits
     awfully
    -2.43
     reap
    -2.38
    Quantity
    -2.23
     Num
    -2.23
     Prepare
    -2.20
     unman
    -2.17
     Participation
    -2.15
     sufficiently
    -2.13
    �
    -2.13
     Perspective
    -2.10
    POSITIVE LOGITS
    Prosecutors
    4.92
     prosecutors
    4.24
    Prosecut
    4.22
     Prosecutors
    4.20
    Judge
    4.11
    Court
    4.11
     prosecutor
    3.94
     COURT
    3.82
     jurors
    3.75
    Detect
    3.69
    Activations Density 0.028%

    No Known Activations