© 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
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 9-ATT-KK
    6. 21151
    Prev
    Next
    INDEX
    Explanations

    phrases that highlight challenges related to criminal justice and public safety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_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.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.06
    1:0.04
    2:0.11
    3:0.20
    4:0.04
    5:0.10
    6:0.03
    7:0.06
    8:0.05
    9:0.03
    10:0.07
    11:0.15
    Negative Logits
     insulted
    -2.74
     amused
    -2.50
     apologised
    -2.42
     remembrance
    -2.37
     pretended
    -2.34
     accuser
    -2.33
     humiliated
    -2.32
     honoured
    -2.31
     despised
    -2.29
     mocked
    -2.25
    POSITIVE LOGITS
    �
    3.75
    [/
    3.55
     �
    3.38
    �
    3.28
    »
    3.27
    ­
    3.09
    ♦
    3.07
    ~~
    3.00
    ═
    2.94
    ====
    2.94
    Activations Density 0.585%

    No Known Activations