© 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. 41492
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names of locations and events

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

    No Comments

    Head Attr Weights
    0:0.02
    1:0.01
    2:0.10
    3:0.04
    4:0.04
    5:0.02
    6:0.41
    7:0.09
    8:0.03
    9:0.05
    10:0.11
    11:0.03
    Negative Logits
    qv
    -1.46
    nown
    -1.42
    ighting
    -1.40
    opsis
    -1.39
    oldown
    -1.38
     Expand
    -1.35
    tnc
    -1.34
    pell
    -1.33
     Karn
    -1.29
     Recomm
    -1.29
    POSITIVE LOGITS
    �士
    1.46
    NFL
    1.38
    Chair
    1.38
    �
    1.26
     reins
    1.24
    士
    1.22
    querque
    1.20
    OTUS
    1.20
    ADA
    1.17
     ("
    1.17
    Activations Density 0.092%

    No Known Activations