© 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. 3908
    Prev
    Next
    INDEX
    Explanations

    expressions of wealth and privilege, particularly in a critical context

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.02
    1:0.02
    2:0.06
    3:0.09
    4:0.13
    5:0.02
    6:0.42
    7:0.06
    8:0.03
    9:0.03
    10:0.04
    11:0.04
    Negative Logits
     margins
    -1.43
     warrants
    -1.43
     similarity
    -1.41
     overlap
    -1.36
     similarities
    -1.32
     margin
    -1.31
     flair
    -1.28
     parallels
    -1.26
     markings
    -1.25
     veins
    -1.23
    POSITIVE LOGITS
    ouver
    1.98
    amed
    1.86
    quished
    1.72
    lished
    1.72
    ammed
    1.65
    owship
    1.65
    successful
    1.65
    uzz
    1.65
    achable
    1.65
    avior
    1.64
    Activations Density 0.168%

    No Known Activations