© 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. 30133
    Prev
    Next
    INDEX
    Explanations

    past tense verbs that suggest actions or events

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    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
    Head Attr Weights
    0:0.17
    1:0.16
    2:0.06
    3:0.05
    4:0.04
    5:0.17
    6:0.04
    7:0.03
    8:0.05
    9:0.07
    10:0.07
    11:0.05
    Negative Logits
     underestimate
    -1.84
    obyl
    -1.83
     underest
    -1.79
    ã
    -1.73
     sqor
    -1.73
    arde
    -1.72
    afort
    -1.71
     hypoc
    -1.69
     overest
    -1.68
     sudden
    -1.68
    POSITIVE LOGITS
    ographs
    1.62
    Georg
    1.60
    igans
    1.58
    Card
    1.54
    ICS
    1.52
     Akira
    1.51
     Recipes
    1.47
     AJ
    1.46
    operative
    1.45
     Kev
    1.45
    Activations Density 0.001%

    No Known Activations