© 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. 9-ATT-KK
    11. 5546
    Prev
    Next
    INDEX
    Explanations

    concepts related to potential consequences and moral implications of decisions

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.07
    1:0.02
    2:0.04
    3:0.18
    4:0.02
    5:0.07
    6:0.02
    7:0.05
    8:0.02
    9:0.01
    10:0.43
    11:0.02
    Negative Logits
     wishing
    -2.18
     embodiments
    -1.98
     Vict
    -1.97
    ITNESS
    -1.97
     renewed
    -1.96
     witnessing
    -1.95
     congrat
    -1.94
     congratulations
    -1.94
     success
    -1.93
     fame
    -1.92
    POSITIVE LOGITS
     altogether
    2.86
     entirely
    2.63
     unnecessary
    2.52
     drastically
    2.52
     wasteful
    2.27
     redund
    2.15
     ineffective
    2.13
    fusc
    2.12
     negligible
    2.04
     useless
    2.02
    Activations Density 0.100%

    No Known Activations