© 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. 3-ATT-KK
    11. 13061
    Prev
    Next
    INDEX
    Explanations

    descriptions of criminal activities and police incidents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    sentences that contain events or actions related to law enforcement or crime

    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
    0:0.04
    1:0.07
    2:0.13
    3:0.04
    4:0.07
    5:0.09
    6:0.09
    7:0.10
    8:0.10
    9:0.06
    10:0.07
    11:0.08
    Negative Logits
     spoiler
    -1.06
     nerds
    -0.99
     paras
    -0.96
     shall
    -0.96
     fucking
    -0.95
     nom
    -0.94
     silly
    -0.93
     shitty
    -0.93
     Philos
    -0.92
     gh
    -0.91
    POSITIVE LOGITS
    eka
    1.16
    anton
    1.10
    ullivan
    1.09
    ğ
    1.08
    oppers
    1.06
    lication
    1.03
    dullah
    1.03
    qqa
    1.02
     spokeswoman
    1.01
    andowski
    1.00
    Activations Density 0.060%

    No Known Activations