© 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. 12549
    Prev
    Next
    INDEX
    Explanations

    important information or emphasis in a text

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

    phrases indicating stress or concern about specific issues

    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
    1:0.03
    2:0.10
    3:0.07
    4:0.02
    5:0.04
    6:0.04
    7:0.08
    8:0.11
    9:0.20
    10:0.07
    11:0.17
    Negative Logits
     gal
    -1.21
    DragonMagazine
    -1.20
    mage
    -1.18
    Gall
    -1.06
    Mill
    -1.02
     cz
    -1.00
    hold
    -1.00
    pires
    -0.99
     Finder
    -0.99
    ea
    -0.99
    POSITIVE LOGITS
     refrain
    1.31
     vigilance
    1.30
     importance
    1.21
     emphasize
    1.20
     stressing
    1.16
     misunderstand
    1.15
    clus
    1.14
    iculture
    1.14
     themes
    1.14
     emphasizing
    1.14
    Activations Density 0.006%

    No Known Activations