© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 10-ATT-KK
    11. 10707
    Prev
    Next
    INDEX
    Explanations

    legal penalties and offenses associated with various actions

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.04
    1:0.03
    2:0.04
    3:0.14
    4:0.08
    5:0.14
    6:0.05
    7:0.14
    8:0.04
    9:0.13
    10:0.05
    11:0.07
    Negative Logits
     Miracle
    -2.19
    ecause
    -2.01
     because
    -2.01
     faked
    -1.99
     Moments
    -1.96
    SPONSORED
    -1.94
    iman
    -1.93
     moments
    -1.90
    cious
    -1.90
     referee
    -1.83
    POSITIVE LOGITS
     +=
    2.56
    】
    2.52
    =(
    2.24
    \-
    2.19
    ,—
    2.19
    vous
    2.18
     ``(
    2.17
    :-
    2.17
    '."
    2.13
    }}
    2.10
    Activations Density 0.014%

    No Known Activations