© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 3-ATT-KK
    11. 14160
    Prev
    Next
    INDEX
    Explanations

    sports-related information, specifically about achievements, changes in coaching staff, and injuries suffered by athletes

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

    entities associated with personal injuries and their related announcements in the news

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.01
    1:0.03
    2:0.08
    3:0.26
    4:0.02
    5:0.03
    6:0.05
    7:0.07
    8:0.08
    9:0.14
    10:0.07
    11:0.12
    Negative Logits
    opa
    -1.42
    ratom
    -1.22
    jit
    -1.19
    eport
    -1.17
    anwhile
    -1.16
    umb
    -1.12
    nai
    -1.10
    utic
    -1.09
    omnia
    -1.08
    nis
    -1.08
    POSITIVE LOGITS
    >]
    1.21
     seal
    1.19
     CRC
    1.10
     -->
    1.09
    sheets
    1.09
     contract
    1.08
     stride
    1.04
     equation
    1.03
    Matrix
    1.02
    ################
    1.01
    Activations Density 0.008%

    No Known Activations