© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 3-ATT-KK
    6. 10656
    Prev
    Next
    INDEX
    Explanations

    instances of legal terms and phrases related to criminal charges and convictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.01
    1:0.07
    2:0.07
    3:0.06
    4:0.02
    5:0.05
    6:0.05
    7:0.10
    8:0.09
    9:0.17
    10:0.11
    11:0.15
    Negative Logits
    DragonMagazine
    -1.24
    Reviewed
    -1.23
    bytes
    -1.16
    strength
    -1.16
    ivot
    -1.13
     endif
    -1.13
    lean
    -1.12
    optim
    -1.10
    actionDate
    -1.09
    ande
    -1.08
    POSITIVE LOGITS
     wrongful
    1.40
     misdemeanor
    1.31
     manslaughter
    1.31
     misdem
    1.26
     involuntary
    1.25
     improper
    1.25
     felony
    1.23
     violation
    1.16
     inconven
    1.15
     tresp
    1.14
    Activations Density 0.007%

    No Known Activations