© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 34226
    Prev
    Next
    INDEX
    Explanations

    references to health risks and exposure to harmful substances

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.02
    1:0.01
    2:0.11
    3:0.07
    4:0.29
    5:0.04
    6:0.05
    7:0.15
    8:0.03
    9:0.04
    10:0.07
    11:0.08
    Negative Logits
    ucket
    -1.61
    */(
    -1.56
    abet
    -1.54
    erto
    -1.42
    Missing
    -1.42
    Vel
    -1.41
    adle
    -1.40
    ));
    -1.38
    istan
    -1.37
    %).
    -1.35
    POSITIVE LOGITS
     hars
    1.51
    erity
    1.46
     goodness
    1.34
     openings
    1.33
     consequences
    1.32
     equivalents
    1.31
     ultraviolet
    1.30
     Almighty
    1.30
     stimuli
    1.30
     sanctioned
    1.29
    Activations Density 0.013%

    No Known Activations