Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 3-ATT-KK
    6. 8591
    Prev
    Next
    INDEX
    Explanations

    references to individuals named Bill, particularly Bill Clinton

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.06
    1:0.04
    2:0.12
    3:0.06
    4:0.04
    5:0.03
    6:0.25
    7:0.14
    8:0.04
    9:0.04
    10:0.09
    11:0.05
    Negative Logits
    ymm
    -1.22
    arna
    -1.14
    angular
    -1.14
    ibaba
    -1.03
    yip
    -1.02
    ogeneity
    -1.01
    undle
    -1.01
    insula
    -0.96
    atl
    -0.95
     ›
    -0.94
    POSITIVE LOGITS
     himself
    1.10
     widow
    1.02
    ichick
    1.01
     brim
    0.99
     baptized
    0.97
     hypocr
    0.97
     vetoed
    0.94
     smoked
    0.92
     silenced
    0.91
     laced
    0.91
    Activations Density 0.029%

    No Known Activations