© 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. 3-ATT-KK
    11. 14318
    Prev
    Next
    INDEX
    Explanations

    informative statements and instructions

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    phrases expressing capability or potential actions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.03
    1:0.08
    2:0.10
    3:0.07
    4:0.02
    5:0.04
    6:0.20
    7:0.07
    8:0.06
    9:0.10
    10:0.08
    11:0.09
    Negative Logits
    EStreamFrame
    -1.15
    accompan
    -0.93
     pol
    -0.93
    OTUS
    -0.89
    pid
    -0.89
    ONSORED
    -0.88
     donor
    -0.88
    isc
    -0.87
     Rober
    -0.84
    plementation
    -0.83
    POSITIVE LOGITS
     yourselves
    1.15
     WATCHED
    1.12
     yourself
    1.10
     Yourself
    1.04
    azaar
    1.03
    veland
    1.02
     Loft
    1.01
     Markus
    1.00
    anamo
    0.99
    ierrez
    0.97
    Activations Density 0.035%

    No Known Activations