© 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. 20608
    Prev
    Next
    INDEX
    Explanations

    references to political events and figures

    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
    0:0.02
    1:0.02
    2:0.03
    3:0.06
    4:0.08
    5:0.45
    6:0.03
    7:0.03
    8:0.02
    9:0.18
    10:0.02
    11:0.03
    Negative Logits
     constantly
    -3.95
     periodically
    -3.81
     annually
    -3.69
     perpetually
    -3.55
     frequently
    -3.50
     regularly
    -3.47
     lately
    -3.40
     daily
    -3.33
     yearly
    -3.19
     every
    -3.18
    POSITIVE LOGITS
     Was
    3.89
     saw
    3.76
    was
    3.66
     consisted
    3.49
    Was
    3.40
     gave
    3.38
     was
    3.35
     showed
    3.29
     WAS
    3.27
     yielded
    3.24
    Activations Density 0.250%

    No Known Activations