© 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. 10886
    Prev
    Next
    INDEX
    Explanations

    phrases that express strong emotions or sentiments, particularly related to negative experiences or moral dilemmas

    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
    1:0.06
    2:0.30
    3:0.06
    4:0.06
    5:0.10
    6:0.03
    7:0.11
    8:0.03
    9:0.04
    10:0.05
    11:0.05
    Negative Logits
     Pengu
    -3.32
     RD
    -3.15
     RM
    -3.05
     BD
    -2.94
    urious
    -2.79
    VP
    -2.78
    PB
    -2.77
     RL
    -2.77
    RM
    -2.76
     PF
    -2.61
    POSITIVE LOGITS
     shall
    7.03
    shall
    6.43
     SHALL
    5.91
     Shall
    5.45
     shalt
    4.74
    NetMessage
    3.63
     thy
    3.60
     "$:/
    3.48
     thee
    3.12
     hereby
    2.98
    Activations Density 0.004%

    No Known Activations