© 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. 24874
    Prev
    Next
    INDEX
    Explanations

    references to "straw man" arguments or concepts indicative of weak reasoning

    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
    0:0.03
    1:0.03
    2:0.06
    3:0.08
    4:0.13
    5:0.05
    6:0.03
    7:0.30
    8:0.06
    9:0.06
    10:0.07
    11:0.04
    Negative Logits
     Provided
    -1.63
    unts
    -1.57
     Universities
    -1.57
    Urban
    -1.54
    owment
    -1.49
    erential
    -1.48
     supplemented
    -1.48
     respected
    -1.47
     Reserved
    -1.44
     livest
    -1.43
    POSITIVE LOGITS
     accusations
    1.86
     setbacks
    1.70
     grievances
    1.70
     allegations
    1.70
     accusation
    1.69
     puzz
    1.68
     cynical
    1.68
     rupture
    1.63
     pitfalls
    1.62
     disappointment
    1.58
    Activations Density 0.000%

    No Known Activations