© 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. 40503
    Prev
    Next
    INDEX
    Explanations

    references to government and regulatory agencies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    0:0.08
    1:0.07
    2:0.03
    3:0.05
    4:0.03
    5:0.21
    6:0.07
    7:0.02
    8:0.06
    9:0.10
    10:0.10
    11:0.11
    Negative Logits
    iaries
    -1.77
     Dates
    -1.53
    jug
    -1.45
    apsed
    -1.30
    ilial
    -1.24
    leep
    -1.22
    packages
    -1.22
    otypes
    -1.21
    estyles
    -1.20
    doms
    -1.20
    POSITIVE LOGITS
     nowhere
    1.43
     distract
    1.26
     Athena
    1.25
     Geoff
    1.25
     vou
    1.14
     Raphael
    1.12
     incess
    1.12
     muted
    1.09
     feedback
    1.07
     constantly
    1.06
    Activations Density 0.037%

    No Known Activations