© 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. 20693
    Prev
    Next
    INDEX
    Explanations

    relationships and connections between concepts, particularly in relation to drawing and its implications

    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
    0:0.20
    1:0.06
    2:0.13
    3:0.03
    4:0.06
    5:0.02
    6:0.10
    7:0.18
    8:0.02
    9:0.03
    10:0.06
    11:0.05
    Negative Logits
     Julian
    -2.75
     Scientology
    -2.65
     Zimmer
    -2.62
     sshd
    -2.52
     Hayden
    -2.50
     systemd
    -2.49
     Scient
    -2.47
     Enterprise
    -2.40
     Damon
    -2.38
     Hayward
    -2.36
    POSITIVE LOGITS
     draw
    5.40
     Draw
    5.36
    Draw
    5.12
     drawing
    5.09
    draw
    4.59
     Drawn
    4.44
     draws
    4.32
     Drawing
    4.32
     drawn
    4.32
     drew
    4.17
    Activations Density 0.035%

    No Known Activations