© 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. 8352
    Prev
    Next
    INDEX
    Explanations

    references to legislative processes and regulations regarding personal rights and medical choices

    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
    0:0.06
    1:0.04
    2:0.07
    3:0.21
    4:0.02
    5:0.09
    6:0.02
    7:0.06
    8:0.03
    9:0.02
    10:0.30
    11:0.02
    Negative Logits
     bewild
    -2.33
     baffled
    -2.25
     puzzled
    -2.24
     aback
    -2.22
     trembling
    -2.22
     dismay
    -2.18
     astonished
    -2.15
    avascript
    -2.15
    �
    -2.10
     unaware
    -2.09
    POSITIVE LOGITS
     permissible
    2.57
     allowable
    2.44
     treats
    2.43
     acceptable
    2.36
     chooses
    2.33
     behaves
    2.33
     SHOULD
    2.29
     stays
    2.26
     should
    2.14
    appropriate
    2.12
    Activations Density 0.125%

    No Known Activations