© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 28237
    Prev
    Next
    INDEX
    Explanations

    phrases involving being cut off or disconnected from something important or essential

    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
    2:0.06
    3:0.08
    4:0.17
    5:0.02
    6:0.03
    7:0.37
    8:0.03
    9:0.03
    10:0.07
    11:0.06
    Negative Logits
    wcsstore
    -1.79
    ivot
    -1.77
    ilts
    -1.67
     toget
    -1.58
    ovember
    -1.55
    estyles
    -1.55
    irtual
    -1.53
    agate
    -1.48
    oln
    -1.47
    oy
    -1.44
    POSITIVE LOGITS
    essential
    1.61
     operator
    1.55
     telephone
    1.54
     supply
    1.51
     nerves
    1.50
     WiFi
    1.46
     nearest
    1.45
     altogether
    1.44
     cable
    1.42
     Tube
    1.42
    Activations Density 0.005%

    No Known Activations