© 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. 12608
    Prev
    Next
    INDEX
    Explanations

    the names of banks and specific individuals

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

    references to specific individuals and organizations involved in various contexts

    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
    1:0.11
    2:0.13
    3:0.02
    4:0.02
    5:0.03
    6:0.11
    7:0.06
    8:0.23
    9:0.04
    10:0.09
    11:0.07
    Negative Logits
    wcs
    -1.47
    illary
    -1.25
    ventions
    -1.24
    headers
    -1.22
    auri
    -1.22
     galvan
    -1.17
    adium
    -1.16
    reddits
    -1.12
    arnaev
    -1.08
     lawy
    -1.06
    POSITIVE LOGITS
    edly
    1.59
     dearly
    1.26
    burgh
    1.21
    ELY
    1.18
     deems
    1.16
    fer
    1.12
    DOC
    1.12
    produ
    1.09
     exec
    1.08
    ifies
    1.06
    Activations Density 0.097%

    No Known Activations