© 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. 14517
    Prev
    Next
    INDEX
    Explanations

    programming-related keywords and code snippets

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    conditional statements and requirements related to specific classifications or rules

    oai_token-act-pair · gpt-4o-mini
    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
    0:0.05
    1:0.03
    2:0.06
    3:0.04
    4:0.06
    5:0.07
    6:0.11
    7:0.32
    8:0.04
    9:0.03
    10:0.08
    11:0.04
    Negative Logits
     Toledo
    -1.15
     Barbie
    -1.11
     Eleanor
    -1.10
     Bohem
    -1.07
    ril
    -1.03
     Conan
    -1.02
     Anita
    -1.01
     Kavanaugh
    -1.01
     childbirth
    -0.99
     chair
    -0.99
    POSITIVE LOGITS
    Downloadha
    1.58
     grouped
    1.39
    ortium
    1.36
    etsk
    1.36
     themselves
    1.31
     individually
    1.25
    edIn
    1.24
     originated
    1.23
     apiece
    1.22
    xtap
    1.20
    Activations Density 0.735%

    No Known Activations