© 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. 14246
    Prev
    Next
    INDEX
    Explanations

    references to specific dates, locations, and names related to news events or historical contexts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    references to significant events or elements related to photography and media

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.03
    1:0.06
    2:0.16
    3:0.03
    4:0.01
    5:0.03
    6:0.08
    7:0.05
    8:0.06
    9:0.06
    10:0.06
    11:0.30
    Negative Logits
     alike
    -1.32
     broadcasters
    -1.24
    venants
    -1.23
    utenberg
    -1.23
    inders
    -1.19
    izons
    -1.13
    iets
    -1.11
    krit
    -1.11
     conferences
    -1.09
     bowls
    -1.09
    POSITIVE LOGITS
     NOTICE
    1.29
     belongs
    1.18
     applies
    1.15
     MUST
    1.12
     titled
    1.09
     assumes
    1.06
     shalt
    1.04
    isphere
    1.04
    CLUS
    1.02
     snippet
    1.02
    Activations Density 0.239%

    No Known Activations