© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 39841
    Prev
    Next
    INDEX
    Explanations

    repeated mentions of the word "the."

    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
    0:0.04
    1:0.03
    2:0.16
    3:0.23
    4:0.08
    5:0.05
    6:0.04
    7:0.02
    8:0.18
    9:0.03
    10:0.03
    11:0.06
    Negative Logits
    htaking
    -2.02
    gettable
    -1.73
    ocally
    -1.70
    ikini
    -1.61
    fits
    -1.58
    emn
    -1.58
    orously
    -1.54
    �
    -1.53
    iera
    -1.53
    anmar
    -1.47
    POSITIVE LOGITS
     Flavoring
    1.90
    田
    1.74
    UA
    1.63
    nexus
    1.55
    �
    1.51
    オ
    1.49
     aug
    1.41
     chances
    1.41
     Logo
    1.39
    anges
    1.38
    Activations Density 0.006%

    No Known Activations