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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 2-ATT-KK
    6. 22311
    Prev
    Next
    INDEX
    Explanations

    references to the "Bald" and "rig" (referring to the "Baldridge Performance Excellence Program")

    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
    0:0.07
    1:0.03
    2:0.24
    3:0.08
    4:0.17
    5:0.05
    6:0.03
    7:0.02
    8:0.07
    9:0.10
    10:0.05
    11:0.02
    Negative Logits
    isSpecialOrderable
    -1.46
    verning
    -1.39
     WHERE
    -1.27
     includ
    -1.22
    groups
    -1.19
     EQ
    -1.18
    accompan
    -1.18
     Cosponsors
    -1.15
    soDeliveryDate
    -1.14
    origin
    -1.12
    POSITIVE LOGITS
     DeV
    1.29
     Poe
    1.27
     Dahl
    1.17
     Schne
    1.16
     Daw
    1.15
     裏�
    1.15
    baugh
    1.15
    SourceFile
    1.13
     Haw
    1.13
     Jr
    1.12
    Activations Density 0.001%

    No Known Activations