© 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. 30240
    Prev
    Next
    INDEX
    Explanations

    negative responses or mentions of closed-door situations

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    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
    0:0.03
    1:0.02
    2:0.05
    3:0.04
    4:0.04
    5:0.03
    6:0.44
    7:0.14
    8:0.04
    9:0.03
    10:0.05
    11:0.04
    Negative Logits
    ashtra
    -1.61
    ollah
    -1.50
    unia
    -1.47
    ��
    -1.46
    umbo
    -1.39
     Wonders
    -1.37
    覚醒
    -1.35
    iaz
    -1.31
    nai
    -1.31
    oulos
    -1.26
    POSITIVE LOGITS
     doors
    1.41
     curtains
    1.41
     door
    1.38
     unpop
    1.33
     thirds
    1.29
     blinded
    1.29
    motion
    1.27
     centimeters
    1.23
     borders
    1.19
    skinned
    1.17
    Activations Density 0.004%

    No Known Activations