© 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. 8956
    Prev
    Next
    INDEX
    Explanations

    specific legal terms and phrases related to allegations and complaints

    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
    1:0.02
    2:0.07
    3:0.03
    4:0.06
    5:0.03
    6:0.23
    7:0.04
    8:0.05
    9:0.28
    10:0.02
    11:0.03
    Negative Logits
    ð
    -3.98
    vector
    -3.93
     Venom
    -3.87
     vector
    -3.65
     Nig
    -3.65
     aur
    -3.58
    �
    -3.45
    Ven
    -3.37
     mercury
    -3.30
     Laksh
    -3.29
    POSITIVE LOGITS
     Robbins
    11.66
    Rob
    6.21
     Robinson
    5.55
     Robb
    5.37
     Rowe
    5.31
     Merrill
    5.31
     Brooks
    5.21
     Briggs
    5.21
     Rob
    4.82
     Robo
    4.77
    Activations Density 0.002%

    No Known Activations