© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 10-ATT-KK
    11. 16846
    Prev
    Next
    INDEX
    Explanations

    conjunctions, particularly the word "and" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.04
    1:0.12
    2:0.03
    3:0.18
    4:0.07
    5:0.03
    6:0.04
    7:0.26
    8:0.03
    9:0.04
    10:0.03
    11:0.08
    Negative Logits
    ODY
    -2.90
     Morning
    -2.53
    Saturday
    -2.52
     Finished
    -2.43
    INAL
    -2.39
    aturday
    -2.33
     Canad
    -2.33
    ilty
    -2.25
     Siber
    -2.23
     Full
    -2.23
    POSITIVE LOGITS
    rils
    2.95
    rake
    2.91
    trap
    2.81
    raid
    2.74
    rir
    2.72
    ras
    2.70
    rex
    2.65
    ril
    2.65
    rupal
    2.53
     runes
    2.51
    Activations Density 0.002%

    No Known Activations