© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 22382
    Prev
    Next
    INDEX
    Explanations

    topics related to political issues and government actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.08
    1:0.06
    2:0.05
    3:0.19
    4:0.08
    5:0.13
    6:0.05
    7:0.07
    8:0.07
    9:0.03
    10:0.08
    11:0.06
    Negative Logits
    VICE
    -2.08
     magically
    -2.02
     outsider
    -1.97
     rabid
    -1.92
     spontaneously
    -1.90
     rewritten
    -1.89
     foreigner
    -1.80
     homemade
    -1.78
     Taiwanese
    -1.76
     spontaneous
    -1.74
    POSITIVE LOGITS
     Bore
    2.43
     Highlands
    2.37
     Depot
    2.37
     Lakes
    2.31
     Peaks
    2.29
     Cache
    2.29
     Meter
    2.26
     Outer
    2.26
     Levels
    2.16
     Crest
    2.15
    Activations Density 0.124%

    No Known Activations