© 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. 39721
    Prev
    Next
    INDEX
    Explanations

    phrases regarding government actions and their implications

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.02
    1:0.02
    2:0.11
    3:0.33
    4:0.09
    5:0.03
    6:0.02
    7:0.07
    8:0.04
    9:0.04
    10:0.08
    11:0.09
    Negative Logits
    bard
    -1.66
    )</
    -1.56
    Lear
    -1.55
    Original
    -1.53
    )*
    -1.52
    wcsstore
    -1.51
    cember
    -1.49
    cellaneous
    -1.49
    Notes
    -1.44
    Legend
    -1.43
    POSITIVE LOGITS
     overshadow
    1.82
     overshadowed
    1.72
     unfairly
    1.71
     improperly
    1.63
     inadequate
    1.62
     perceived
    1.59
     inappropriately
    1.56
     contradicted
    1.54
     outweigh
    1.52
     amid
    1.52
    Activations Density 0.434%

    No Known Activations