© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 3-ATT-KK
    11. 11994
    Prev
    Next
    INDEX
    Explanations

    phrases related to mapping and organizing data structures

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

    phrases related to mapping and data management processes

    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
    1:0.05
    2:0.11
    3:0.06
    4:0.02
    5:0.04
    6:0.05
    7:0.07
    8:0.11
    9:0.28
    10:0.07
    11:0.09
    Negative Logits
     beware
    -1.40
    mbuds
    -1.32
    Synopsis
    -1.28
     fortunately
    -1.26
    elight
    -1.21
     pumps
    -1.15
     vigilance
    -1.15
     lets
    -1.14
     apologies
    -1.14
     advises
    -1.13
    POSITIVE LOGITS
     disparate
    1.53
    ゼウス
    1.26
     differently
    1.24
     Beir
    1.23
     dots
    1.20
     Gou
    1.19
    JV
    1.18
    soDeliveryDate
    1.16
    tnc
    1.15
     McMaster
    1.13
    Activations Density 0.007%

    No Known Activations