© 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. 3-ATT-KK
    11. 14319
    Prev
    Next
    INDEX
    Explanations

    mentions of condemnation, threats, and rhetoric related to difficult societal topics

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    words associated with condemnation and negative assessments

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.03
    1:0.03
    2:0.09
    3:0.25
    4:0.02
    5:0.03
    6:0.10
    7:0.08
    8:0.05
    9:0.07
    10:0.06
    11:0.13
    Negative Logits
    uthor
    -1.28
     reckoning
    -1.16
    drm
    -1.09
    meta
    -1.04
    hold
    -1.03
    uni
    -1.00
    root
    -0.99
    phabet
    -0.99
     fermentation
    -0.98
     sled
    -0.98
    POSITIVE LOGITS
    ruary
    1.38
    ça
    1.28
    uve
    1.27
    owder
    1.21
    APTER
    1.07
    ctuary
    1.06
     ACTIONS
    1.06
    gence
    1.05
     Veil
    1.04
    terday
    1.03
    Activations Density 0.003%

    No Known Activations