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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 2-ATT-KK
    11. 22890
    Prev
    Next
    INDEX
    Explanations

    words related to technology, innovation, and testing

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

    words and phrases related to testing and evaluation

    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
    0:0.04
    1:0.03
    2:0.20
    3:0.05
    4:0.07
    5:0.08
    6:0.02
    7:0.03
    8:0.12
    9:0.22
    10:0.06
    11:0.03
    Negative Logits
     Ast
    -1.36
     Lat
    -1.16
     Wonderland
    -1.16
     twisting
    -1.11
     denying
    -1.09
     whispering
    -1.09
     Anat
    -1.08
    theless
    -1.06
    ilda
    -1.05
     ashamed
    -1.04
    POSITIVE LOGITS
    =-=-=-=-
    1.21
    olphin
    1.17
     sqor
    1.16
    proven
    1.15
    undown
    1.12
    prep
    1.12
    iets
    1.12
    eatures
    1.12
     Flavoring
    1.11
    ipel
    1.11
    Activations Density 0.013%

    No Known Activations