© 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. 5-ATT-KK
    11. 40618
    Prev
    Next
    INDEX
    Explanations

    phrases indicating emphasis or significance, particularly involving the word "of"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.03
    1:0.01
    2:0.11
    3:0.06
    4:0.12
    5:0.04
    6:0.04
    7:0.32
    8:0.03
    9:0.05
    10:0.08
    11:0.07
    Negative Logits
    wives
    -1.60
    achers
    -1.42
     Teachers
    -1.38
    keys
    -1.34
    mouth
    -1.34
    Law
    -1.34
    common
    -1.33
    Hun
    -1.27
    encers
    -1.27
    ydia
    -1.26
    POSITIVE LOGITS
     arrang
    1.62
    paren
    1.49
     affair
    1.45
     arrangements
    1.41
     engagement
    1.35
     arrangement
    1.34
     avoidance
    1.34
     sidel
    1.34
     EVENTS
    1.34
     approving
    1.32
    Activations Density 0.000%

    No Known Activations