© 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. 23075
    Prev
    Next
    INDEX
    Explanations

    Keywords related to political, economic, and societal news or articles

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

    phrases indicating quantity and intensity 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
    0:0.08
    1:0.04
    2:0.13
    3:0.07
    4:0.07
    5:0.08
    6:0.03
    7:0.03
    8:0.16
    9:0.15
    10:0.06
    11:0.05
    Negative Logits
    soever
    -1.51
    ivation
    -1.45
    SPONSORED
    -1.40
    emouth
    -1.31
    ocument
    -1.31
    Reply
    -1.29
    endment
    -1.29
    adium
    -1.26
    verning
    -1.22
    arget
    -1.19
    POSITIVE LOGITS
     pedest
    1.18
     premie
    1.17
    abre
    1.15
     Belt
    1.14
    Veter
    1.12
    QUI
    1.10
    Laughs
    1.08
    Monster
    1.07
     Eye
    1.07
     Heroes
    1.07
    Activations Density 0.001%

    No Known Activations