© 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. 19533
    Prev
    Next
    INDEX
    Explanations

    definitive statements regarding legal or financial matters

    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
    0:0.23
    1:0.05
    2:0.07
    3:0.06
    4:0.05
    5:0.04
    6:0.08
    7:0.07
    8:0.03
    9:0.06
    10:0.15
    11:0.08
    Negative Logits
     Sheldon
    -3.35
     Morse
    -3.33
    Minecraft
    -3.28
     Caps
    -3.10
     Bitcoins
    -3.10
     implants
    -3.07
     Capitals
    -3.05
     almonds
    -3.02
     pac
    -2.99
     PB
    -2.99
    POSITIVE LOGITS
     Bug
    6.48
    Bug
    5.08
    bug
    4.11
     Hou
    3.89
    BUG
    3.87
     Viper
    3.84
     Tripoli
    3.77
     Bugs
    3.69
     Libyan
    3.52
     Fiat
    3.38
    Activations Density 0.000%

    No Known Activations