© 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. 9-ATT-KK
    11. 18513
    Prev
    Next
    INDEX
    Explanations

    numerical data and references to identifiers or labels

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.08
    1:0.04
    2:0.04
    3:0.18
    4:0.20
    5:0.10
    6:0.05
    7:0.09
    8:0.04
    9:0.04
    10:0.03
    11:0.07
    Negative Logits
    orsi
    -2.76
     disse
    -2.33
    racuse
    -2.27
    nces
    -2.24
    aterasu
    -2.17
     rebutt
    -2.14
     colle
    -2.13
    inse
    -2.12
     crit
    -2.08
     qual
    -2.08
    POSITIVE LOGITS
     CHRIST
    3.13
     BUS
    3.07
    ETH
    2.95
     SERVICE
    2.87
     WHITE
    2.86
     GREEN
    2.82
    FL
    2.82
     MIC
    2.82
     MARK
    2.82
    AMY
    2.82
    Activations Density 0.002%

    No Known Activations