© 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. 2-ATT-KK
    11. 22989
    Prev
    Next
    INDEX
    Explanations

    violence-related language, particularly focusing on physical attacks and assaults

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    instances of the word "ripped" and its variants in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    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
    2:0.14
    3:0.10
    4:0.05
    5:0.06
    6:0.03
    7:0.03
    8:0.22
    9:0.12
    10:0.05
    11:0.03
    Negative Logits
    chenko
    -1.17
    illery
    -1.13
    Coun
    -1.11
     councillor
    -1.11
     Assist
    -1.11
     Strateg
    -1.10
     Levine
    -1.09
    SEA
    -1.09
     commanding
    -1.05
     Adviser
    -1.05
    POSITIVE LOGITS
     pulp
    1.39
    away
    1.37
    ���
    1.37
     Dracula
    1.30
    shit
    1.27
     crap
    1.22
    thro
    1.22
    sheets
    1.20
     bits
    1.20
    unic
    1.17
    Activations Density 0.007%

    No Known Activations