© 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. 3-ATT-KK
    11. 12126
    Prev
    Next
    INDEX
    Explanations

    words related to legal matters or complaints about human rights issues

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

    terms related to critical aspects or pivotal elements in discussions or narratives

    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
    0:0.02
    1:0.04
    2:0.08
    3:0.19
    4:0.02
    5:0.03
    6:0.09
    7:0.09
    8:0.05
    9:0.07
    10:0.05
    11:0.23
    Negative Logits
     Ark
    -1.15
     Shutterstock
    -1.13
     Jehovah
    -1.10
     hereafter
    -1.09
     Barn
    -1.07
    yah
    -1.04
    elf
    -1.04
    slave
    -1.03
     Kok
    -1.03
    atern
    -1.02
    POSITIVE LOGITS
     nail
    1.24
    の�
    1.24
    将
    1.23
    ties
    1.11
    rency
    1.07
    remove
    1.07
     reversal
    1.06
    opsis
    1.06
    cludes
    1.06
    breaker
    1.06
    Activations Density 0.005%

    No Known Activations