© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 2-ATT-KK
    11. 11436
    Prev
    Next
    INDEX
    Explanations

    information related to a crime involving the victim, attackers, and their connection to legal proceedings

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.07
    1:0.05
    2:0.11
    3:0.05
    4:0.05
    5:0.11
    6:0.03
    7:0.04
    8:0.05
    9:0.27
    10:0.07
    11:0.04
    Negative Logits
    soDeliveryDate
    -1.86
     isEnabled
    -1.35
    Marvel
    -1.31
    エ
    -1.23
     billionaires
    -1.21
     caveats
    -1.21
     esports
    -1.20
    isSpecialOrderable
    -1.18
     Games
    -1.17
     antitrust
    -1.17
    POSITIVE LOGITS
     Mara
    1.41
     PTSD
    1.40
     Ramirez
    1.38
    emort
    1.35
     syndrome
    1.28
    quez
    1.24
     untreated
    1.22
    rano
    1.22
     Victim
    1.22
     Rape
    1.21
    Activations Density 0.049%

    No Known Activations