© 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. 9854
    Prev
    Next
    INDEX
    Explanations

    numerical values and formatting related to books

    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
    0:0.20
    1:0.02
    2:0.05
    3:0.13
    4:0.05
    5:0.11
    6:0.03
    7:0.11
    8:0.06
    9:0.03
    10:0.11
    11:0.04
    Negative Logits
     moot
    -2.13
     subreddits
    -1.98
    76561
    -1.97
     broadcasters
    -1.97
     broadcasting
    -1.90
     reminis
    -1.89
     intakes
    -1.86
    ury
    -1.84
     telev
    -1.83
    wives
    -1.81
    POSITIVE LOGITS
     paperback
    3.16
     ISBN
    3.08
    Paper
    2.72
     Pages
    2.56
     Kindle
    2.48
    printed
    2.43
     Publication
    2.41
     COVER
    2.39
     sleeve
    2.36
     ebook
    2.28
    Activations Density 0.010%

    No Known Activations