© 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. 3-ATT-KK
    11. 5274
    Prev
    Next
    INDEX
    Explanations

    phrases related to making apologies or seeking forgiveness

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.04
    1:0.03
    2:0.08
    3:0.04
    4:0.02
    5:0.07
    6:0.05
    7:0.21
    8:0.08
    9:0.06
    10:0.09
    11:0.17
    Negative Logits
    glomer
    -1.58
    ixture
    -1.41
    virtual
    -1.30
    ghai
    -1.28
    unin
    -1.27
    onomy
    -1.24
    asus
    -1.22
    urated
    -1.22
    ossibility
    -1.21
     specialization
    -1.19
    POSITIVE LOGITS
    Reward
    1.25
     gag
    1.24
    �醒
    1.22
     apology
    1.21
     grat
    1.19
     toile
    1.18
    jured
    1.18
    ovember
    1.15
     sore
    1.13
     Pledge
    1.13
    Activations Density 0.017%

    No Known Activations