© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 3-ATT-KK
    11. 14473
    Prev
    Next
    INDEX
    Explanations

    prepositions and nouns associated with a location or position

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    instances of the word "running" in various contexts or applications

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Head Attr Weights
    0:0.01
    1:0.02
    2:0.14
    3:0.16
    4:0.02
    5:0.02
    6:0.06
    7:0.12
    8:0.08
    9:0.18
    10:0.05
    11:0.06
    Negative Logits
    etheless
    -1.01
    lege
    -0.95
    mean
    -0.94
    Crit
    -0.93
    telling
    -0.93
    ーテ
    -0.93
    orative
    -0.93
    eeks
    -0.93
     thanking
    -0.93
    fecture
    -0.92
    POSITIVE LOGITS
    interstitial
    1.58
    thumbnails
    1.32
     mirrors
    1.25
    Downloadha
    1.11
     Impossible
    1.09
    wagon
    1.09
     curtains
    1.09
    ��
    1.07
    gradient
    1.05
    doors
    1.04
    Activations Density 0.020%

    No Known Activations