© 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. 37418
    Prev
    Next
    INDEX
    Explanations

    instances of the word "the" and variations of the word "it"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    1:0.01
    2:0.06
    3:0.12
    4:0.18
    5:0.03
    6:0.03
    7:0.30
    8:0.03
    9:0.04
    10:0.06
    11:0.08
    Negative Logits
     tracts
    -1.29
     Afric
    -1.27
    ワ
    -1.23
    itles
    -1.22
    notes
    -1.21
    cloneembedreportprint
    -1.19
    Americ
    -1.19
     Yep
    -1.18
    ominated
    -1.17
     kernels
    -1.16
    POSITIVE LOGITS
    loo
    1.63
     abrupt
    1.51
     inaction
    1.43
     interference
    1.36
     deterioration
    1.33
     bureaucracy
    1.32
     intrusion
    1.30
     Boss
    1.27
     wasteful
    1.23
     Cruel
    1.22
    Activations Density 0.006%

    No Known Activations