© 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. 14122
    Prev
    Next
    INDEX
    Explanations

    words related to fundamental characteristics or essential qualities

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

    ideas related to inherent qualities or risks associated with various concepts or entities

    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
    0:0.02
    1:0.02
    2:0.07
    3:0.12
    4:0.02
    5:0.04
    6:0.09
    7:0.14
    8:0.05
    9:0.08
    10:0.08
    11:0.22
    Negative Logits
    apest
    -1.43
    osponsors
    -1.26
    opausal
    -1.21
    llor
    -1.19
    faced
    -1.15
    eday
    -1.14
    bye
    -1.10
    arger
    -1.07
    armac
    -1.05
    iversary
    -1.05
    POSITIVE LOGITS
    �
    1.32
     relativity
    1.10
    ーテ
    1.06
     Unity
    1.06
    abilities
    1.06
    gravity
    1.05
     equation
    1.05
     worldview
    1.05
     sense
    1.04
     physics
    1.04
    Activations Density 0.007%

    No Known Activations