© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 45172
    Prev
    Next
    INDEX
    Explanations

    references to functions in mathematics, particularly rational functions

    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
    0:0.07
    1:0.06
    2:0.09
    3:0.08
    4:0.08
    5:0.07
    6:0.07
    7:0.08
    8:0.08
    9:0.09
    10:0.10
    11:0.07
    Negative Logits
    atre
    -1.65
    illard
    -1.63
     presided
    -1.57
    %;
    -1.55
    %]
    -1.55
     Dug
    -1.51
     colleg
    -1.49
     Nadu
    -1.48
     ⓘ
    -1.48
     Beir
    -1.48
    POSITIVE LOGITS
    raq
    2.30
    ilts
    1.89
    ocument
    1.80
    CVE
    1.74
    zees
    1.73
    NES
    1.70
    eas
    1.70
    Trigger
    1.70
    PLIED
    1.67
    Redditor
    1.66
    Activations Density 0.000%

    No Known Activations