© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 5-ATT-KK
    11. 36352
    Prev
    Next
    INDEX
    Explanations

    significant mentions of the word "the" and its context in discussions about governmental or organizational actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    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
    0:0.02
    1:0.02
    2:0.06
    3:0.06
    4:0.24
    5:0.05
    6:0.06
    7:0.18
    8:0.05
    9:0.04
    10:0.08
    11:0.08
    Negative Logits
    tips
    -1.58
    naire
    -1.58
     emot
    -1.46
    iddles
    -1.45
    Tips
    -1.45
    ience
    -1.44
    illin
    -1.44
    Redditor
    -1.42
    ummies
    -1.42
    sle
    -1.41
    POSITIVE LOGITS
     stronghold
    1.50
     Presidency
    1.50
    omore
    1.47
     garrison
    1.43
     inauguration
    1.42
     pledge
    1.37
     Brune
    1.33
     compound
    1.31
     Mountain
    1.30
    halla
    1.29
    Activations Density 0.001%

    No Known Activations