© 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. 4079
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "autop" and its variations, reflecting a focus on autonomous technologies or systems

    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
    2:0.04
    3:0.06
    4:0.05
    5:0.05
    6:0.43
    7:0.03
    8:0.05
    9:0.07
    10:0.07
    11:0.05
    Negative Logits
     proxies
    -1.46
     wid
    -1.38
     orphans
    -1.38
     handwriting
    -1.29
     hypoc
    -1.27
     chores
    -1.25
     proxy
    -1.24
     wide
    -1.24
    FUL
    -1.24
     boundaries
    -1.21
    POSITIVE LOGITS
    lycer
    1.75
    ideo
    1.69
    enza
    1.63
    ciating
    1.61
    ronic
    1.58
    ét
    1.52
    anium
    1.49
    cus
    1.48
    incial
    1.48
    ufact
    1.48
    Activations Density 0.004%

    No Known Activations