© 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. 9-ATT-KK
    11. 10948
    Prev
    Next
    INDEX
    Explanations

    conjunctions and punctuation, particularly focusing on the word 'and' as well as various forms of commas and semicolons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.05
    1:0.15
    2:0.06
    3:0.07
    4:0.07
    5:0.05
    6:0.06
    7:0.21
    8:0.04
    9:0.07
    10:0.08
    11:0.04
    Negative Logits
    Enough
    -2.57
    Different
    -2.43
     precon
    -2.30
    usha
    -2.30
    There
    -2.29
    du
    -2.29
    Orig
    -2.25
    Native
    -2.21
    Assuming
    -2.17
    Actually
    -2.15
    POSITIVE LOGITS
    ilial
    3.14
    mercial
    2.84
    then
    2.79
    illon
    2.73
     Shib
    2.72
    odied
    2.64
    later
    2.61
    second
    2.55
     Tackle
    2.55
    yles
    2.52
    Activations Density 0.001%

    No Known Activations