© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 2-ATT-KK
    11. 23008
    Prev
    Next
    INDEX
    Explanations

    government agencies, regulations, law enforcement activities, and reports on various investigations and statistics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    phrases referring to various U.S. government agencies and their functions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    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
    1:0.03
    2:0.16
    3:0.05
    4:0.30
    5:0.05
    6:0.02
    7:0.02
    8:0.07
    9:0.14
    10:0.04
    11:0.01
    Negative Logits
    aways
    -1.61
     peripher
    -1.49
    ours
    -1.46
     :-)
    -1.46
    castle
    -1.45
    stre
    -1.43
    azer
    -1.42
     🙂
    -1.40
    ouched
    -1.38
    pert
    -1.38
    POSITIVE LOGITS
     Forestry
    1.63
    rehens
    1.62
     Corrections
    1.61
     Reconstruction
    1.58
     Superintendent
    1.56
    ット
    1.55
     Bureau
    1.52
     Discipline
    1.50
     Doctrine
    1.49
     Examination
    1.47
    Activations Density 0.007%

    No Known Activations