© 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. 9-ATT-KK
    11. 15508
    Prev
    Next
    INDEX
    Explanations

    phrases that express conditional or contradicted beliefs and complex arguments

    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
    0:0.09
    1:0.02
    2:0.08
    3:0.20
    4:0.02
    5:0.07
    6:0.02
    7:0.05
    8:0.03
    9:0.02
    10:0.33
    11:0.02
    Negative Logits
     arrival
    -2.47
     disembark
    -1.94
     deployment
    -1.92
     exodus
    -1.91
     landing
    -1.80
     situation
    -1.79
     sighting
    -1.77
    patch
    -1.76
     arriving
    -1.74
     signing
    -1.73
    POSITIVE LOGITS
     antit
    2.27
    pires
    2.19
    OULD
    2.14
    pired
    2.11
    harm
    2.08
     undesirable
    2.07
     intrinsically
    2.02
     disliked
    2.00
     beneficial
    1.98
     inherently
    1.94
    Activations Density 0.423%

    No Known Activations