© 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. 5-ATT-KK
    11. 16925
    Prev
    Next
    INDEX
    Explanations

    different instances of the word "to" in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.12
    1:0.14
    2:0.04
    3:0.04
    4:0.03
    5:0.18
    6:0.06
    7:0.06
    8:0.11
    9:0.06
    10:0.05
    11:0.05
    Negative Logits
     lend
    -1.38
    invoke
    -1.34
     testify
    -1.32
    ��
    -1.31
     recall
    -1.30
     Belg
    -1.30
     eman
    -1.28
    film
    -1.25
     derive
    -1.23
     deploy
    -1.22
    POSITIVE LOGITS
    romeda
    1.65
     Solitaire
    1.58
    /?
    1.57
     Shopping
    1.51
    cest
    1.45
    cation
    1.42
     ||
    1.41
    ulo
    1.40
     Hispan
    1.37
    umbn
    1.37
    Activations Density 0.008%

    No Known Activations