© 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. 33369
    Prev
    Next
    INDEX
    Explanations

    references to "the" and other articles, indicating a focus on specific entities or subjects within the text

    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
    0:0.04
    1:0.06
    2:0.05
    3:0.07
    4:0.11
    5:0.11
    6:0.02
    7:0.10
    8:0.10
    9:0.10
    10:0.09
    11:0.09
    Negative Logits
    perse
    -1.85
    ÃÂ
    -1.45
    Downloadha
    -1.40
    ser
    -1.40
    Exper
    -1.39
    �
    -1.38
    Zip
    -1.37
    rose
    -1.35
    esters
    -1.33
    ictionary
    -1.33
    POSITIVE LOGITS
     developments
    1.60
     eventual
    1.57
     bureaucrats
    1.39
     inaction
    1.38
     DeVos
    1.38
     deport
    1.32
     Committee
    1.31
     broader
    1.31
     Peninsula
    1.29
     Soviets
    1.28
    Activations Density 0.122%

    No Known Activations