© 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. 14416
    Prev
    Next
    INDEX
    Explanations

    reasons or explanations within text

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

    phrases indicating assertions or explanations about reasons or facts

    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
    0:0.01
    1:0.09
    2:0.14
    3:0.04
    4:0.01
    5:0.02
    6:0.08
    7:0.14
    8:0.16
    9:0.17
    10:0.04
    11:0.05
    Negative Logits
     avoids
    -1.36
     enjoys
    -1.33
     publishes
    -1.32
     recovers
    -1.27
     develops
    -1.26
     consumes
    -1.25
     wears
    -1.24
     restores
    -1.24
     combines
    -1.24
    skirts
    -1.23
    POSITIVE LOGITS
     Beir
    1.67
    actly
    1.46
    cedented
    1.43
    nah
    1.42
    PLIED
    1.41
    _-
    1.41
     understatement
    1.41
    erial
    1.40
    ighed
    1.38
    omal
    1.37
    Activations Density 0.053%

    No Known Activations