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

    Neuronpedia

    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. 10-ATT-KK
    11. 15247
    Prev
    Next
    INDEX
    Explanations

    discussions about political figures, particularly focusing on their actions and influence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.08
    1:0.38
    2:0.06
    3:0.03
    4:0.03
    5:0.01
    6:0.10
    7:0.09
    8:0.02
    9:0.02
    10:0.05
    11:0.06
    Negative Logits
     Gran
    -2.43
    reddits
    -2.38
    hed
    -2.34
    ader
    -2.30
     Latter
    -2.29
    bian
    -2.29
    ally
    -2.21
    erver
    -2.20
     Eucl
    -2.19
    iday
    -2.18
    POSITIVE LOGITS
     Power
    8.27
    Power
    8.25
     power
    7.57
    power
    7.41
     POWER
    7.11
     Powers
    5.72
     powers
    4.97
    powers
    4.79
     powering
    4.60
    powered
    4.59
    Activations Density 0.196%

    No Known Activations