© 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. 10-ATT-KK
    11. 22854
    Prev
    Next
    INDEX
    Explanations

    mentions of accidents and their consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.05
    1:0.02
    2:0.04
    3:0.10
    4:0.17
    5:0.24
    6:0.03
    7:0.06
    8:0.02
    9:0.18
    10:0.02
    11:0.03
    Negative Logits
    \">
    -3.19
     Consent
    -3.09
    �
    -2.84
     approves
    -2.81
    xual
    -2.80
     distinguishes
    -2.77
     prefers
    -2.76
     Doesn
    -2.71
     invites
    -2.68
     Makes
    -2.64
    POSITIVE LOGITS
     charred
    3.88
     collapsed
    3.73
     engulfed
    3.72
     shattered
    3.65
     toppled
    3.55
     burnt
    3.48
     panicked
    3.47
     smashed
    3.38
     wreckage
    3.36
     burned
    3.35
    Activations Density 0.472%

    No Known Activations