© 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. 38348
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate roles or identities related to individuals in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.01
    1:0.00
    2:0.14
    3:0.08
    4:0.15
    5:0.01
    6:0.03
    7:0.22
    8:0.03
    9:0.02
    10:0.14
    11:0.09
    Negative Logits
    vable
    -1.23
    antine
    -1.10
    benefit
    -1.05
    optional
    -1.05
     Amendments
    -1.04
    Import
    -1.01
     strengthens
    -1.00
    henko
    -0.98
     provision
    -0.98
     ballots
    -0.98
    POSITIVE LOGITS
    龍喚士
    1.41
     solitude
    1.29
     persona
    1.21
     Leilan
    1.20
    WithNo
    1.20
     Frie
    1.20
    isexual
    1.11
    �
    1.11
    irlf
    1.11
    gamer
    1.10
    Activations Density 0.200%

    No Known Activations