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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 20671
    Prev
    Next
    INDEX
    Explanations

    conditional statements and expressions, particularly those that contrast or qualify assertions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    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
    1:0.02
    2:0.08
    3:0.10
    4:0.17
    5:0.04
    6:0.04
    7:0.07
    8:0.11
    9:0.07
    10:0.09
    11:0.11
    Negative Logits
    ヘ
    -1.70
    ゼウス
    -1.43
     discont
    -1.25
    者
    -1.21
    ?'
    -1.15
    ?".
    -1.13
    ラ
    -1.12
    ?'"
    -1.11
    ascus
    -1.10
     assemblies
    -1.10
    POSITIVE LOGITS
    wcsstore
    1.48
    soType
    1.22
    Lago
    1.22
    atsuki
    1.20
    Pract
    1.18
    External
    1.16
    FLAG
    1.15
     Laur
    1.15
    hound
    1.14
     Treasurer
    1.13
    Activations Density 0.044%

    No Known Activations