© 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. 12799
    Prev
    Next
    INDEX
    Explanations

    references to financial costs and budgeting in a political or governmental context

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

    financial terms and metrics related to costs and economic measures

    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
    1:0.01
    2:0.04
    3:0.31
    4:0.01
    5:0.03
    6:0.06
    7:0.16
    8:0.04
    9:0.08
    10:0.09
    11:0.09
    Negative Logits
    ergy
    -1.41
    �
    -1.35
    ////////////////
    -1.31
     Jehovah
    -1.28
    �
    -1.27
    AppData
    -1.24
    ��
    -1.18
    ENN
    -1.16
    xa
    -1.15
    eon
    -1.14
    POSITIVE LOGITS
     incarcer
    1.35
     incurred
    1.33
     estimating
    1.15
     averaging
    1.12
     bailout
    1.11
     dearly
    1.11
     upkeep
    1.10
     assembling
    1.10
     fixing
    1.09
     adjourn
    1.09
    Activations Density 0.030%

    No Known Activations