© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 38118
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly commas

    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
    0:0.06
    1:0.07
    2:0.13
    3:0.06
    4:0.11
    5:0.07
    6:0.03
    7:0.06
    8:0.19
    9:0.04
    10:0.06
    11:0.07
    Negative Logits
    itably
    -1.56
    hor
    -1.48
    hire
    -1.37
     paw
    -1.35
     guess
    -1.35
     y
    -1.31
     bip
    -1.29
     cro
    -1.29
    Car
    -1.28
    hesis
    -1.27
    POSITIVE LOGITS
    cyclopedia
    1.68
    abwe
    1.63
    deen
    1.58
    quartered
    1.56
    etheless
    1.54
    vre
    1.46
    ドラゴン
    1.45
     Mub
    1.43
    vance
    1.43
    phia
    1.41
    Activations Density 0.001%

    No Known Activations