© 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. 2-ATT-KK
    11. 11530
    Prev
    Next
    INDEX
    Explanations

    names mixed with possible fictional or scientific elements

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.08
    1:0.02
    2:0.38
    3:0.08
    4:0.09
    5:0.05
    6:0.02
    7:0.02
    8:0.05
    9:0.08
    10:0.05
    11:0.02
    Negative Logits
     Flavoring
    -1.50
    yip
    -1.49
    Downloadha
    -1.43
    iversal
    -1.35
    monary
    -1.30
    okemon
    -1.29
    DragonMagazine
    -1.27
    ongyang
    -1.23
     ACS
    -1.23
    mercial
    -1.19
    POSITIVE LOGITS
    quez
    1.56
    otte
    1.45
    abeth
    1.44
    ija
    1.30
    igne
    1.30
     Perez
    1.24
     Grant
    1.22
    ado
    1.22
     Musk
    1.22
     Isles
    1.19
    Activations Density 0.020%

    No Known Activations