© 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. 8758
    Prev
    Next
    INDEX
    Explanations

    expressions related to people and their actions or experiences

    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
    2:0.14
    3:0.08
    4:0.05
    5:0.05
    6:0.15
    7:0.04
    8:0.06
    9:0.05
    10:0.15
    11:0.15
    Negative Logits
    cone
    -1.29
    ois
    -1.28
    thing
    -1.21
    outine
    -1.13
    akery
    -1.08
    cube
    -1.04
    osure
    -1.03
     Fork
    -1.03
    iable
    -1.03
     Juliet
    -1.02
    POSITIVE LOGITS
    ali
    1.29
    Redditor
    1.25
     Antar
    1.19
     victimized
    1.19
     suscept
    1.18
    mol
    1.15
     [|
    1.15
     whom
    1.12
    ِ
    1.12
     caption
    1.12
    Activations Density 0.063%

    No Known Activations