© 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. 10625
    Prev
    Next
    INDEX
    Explanations

    instances of emotional impact or significant reactions in situations

    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
    0:0.16
    1:0.10
    2:0.08
    3:0.06
    4:0.05
    5:0.03
    6:0.08
    7:0.07
    8:0.03
    9:0.04
    10:0.17
    11:0.07
    Negative Logits
     Mack
    -3.41
     Huang
    -3.18
     McM
    -2.95
     Frem
    -2.91
     Bundy
    -2.91
     Mick
    -2.87
     LM
    -2.87
    Bey
    -2.86
    Mas
    -2.82
    inance
    -2.81
    POSITIVE LOGITS
     curs
    7.04
     cursor
    4.71
     handwriting
    3.75
    CV
    3.36
     curses
    3.28
     Cros
    3.23
     scrib
    3.14
     CV
    3.10
    Cro
    3.08
    ursor
    3.02
    Activations Density 0.000%

    No Known Activations