© 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. 47246
    Prev
    Next
    INDEX
    Explanations

    statements or phrases that emphasize caution or warnings

    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
    0:0.02
    1:0.01
    2:0.09
    3:0.09
    4:0.12
    5:0.02
    6:0.05
    7:0.39
    8:0.03
    9:0.03
    10:0.04
    11:0.06
    Negative Logits
    cro
    -1.59
    arth
    -1.49
    ros
    -1.35
    jee
    -1.34
    downs
    -1.33
    houses
    -1.32
    lot
    -1.32
    ighth
    -1.30
     renovated
    -1.29
    obin
    -1.28
    POSITIVE LOGITS
     Commit
    1.47
     exagger
    1.41
    incerity
    1.40
     disclaim
    1.37
     Deploy
    1.36
    Narr
    1.34
     accompl
    1.33
     Principles
    1.32
     Guilty
    1.31
     behavi
    1.30
    Activations Density 0.001%

    No Known Activations