© 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. 2-ATT-KK
    11. 11520
    Prev
    Next
    INDEX
    Explanations

    phrases related to the condition or quality of a place, including mentions of staying, living, democracy, and trust

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.04
    1:0.04
    2:0.07
    3:0.12
    4:0.04
    5:0.11
    6:0.02
    7:0.03
    8:0.07
    9:0.29
    10:0.07
    11:0.04
    Negative Logits
    vous
    -1.17
     delay
    -1.14
    alions
    -1.13
     msec
    -1.10
    TPP
    -1.07
    ernaut
    -1.06
    uten
    -1.01
     ppm
    -1.00
    erity
    -1.00
     Delay
    -0.99
    POSITIVE LOGITS
     Ascension
    1.31
    ulia
    1.24
     neighb
    1.11
    yg
    1.11
     Cube
    1.06
    hung
    1.06
     Satanic
    1.06
     Eucl
    1.06
    mite
    1.06
    inois
    1.05
    Activations Density 0.088%

    No Known Activations