© 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. 9-ATT-KK
    11. 4191
    Prev
    Next
    INDEX
    Explanations

    statements expressing difficulty or complexity in tasks

    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
    0:0.08
    1:0.03
    2:0.09
    3:0.08
    4:0.03
    5:0.04
    6:0.03
    7:0.40
    8:0.03
    9:0.03
    10:0.08
    11:0.02
    Negative Logits
     Broadcasting
    -2.46
    Liber
    -2.40
    allah
    -2.38
     Appl
    -2.36
    SourceFile
    -2.32
    rawdownloadcloneembedreportprint
    -2.29
     BaseType
    -2.27
     abundance
    -2.22
    ��
    -2.22
     ItemLevel
    -2.21
    POSITIVE LOGITS
    ascript
    3.40
     difficult
    3.24
     complicated
    3.17
     complexities
    3.02
     tricky
    3.01
    Second
    2.92
     cumbers
    2.89
     complexity
    2.83
     slower
    2.78
     harder
    2.77
    Activations Density 0.048%

    No Known Activations