© 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. 14184
    Prev
    Next
    INDEX
    Explanations

    dates supplemented by event descriptions

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

    guest posts and associated metadata in articles

    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
    1:0.13
    2:0.05
    3:0.04
    4:0.02
    5:0.06
    6:0.06
    7:0.12
    8:0.07
    9:0.13
    10:0.05
    11:0.21
    Negative Logits
     disadvantage
    -1.68
     decay
    -1.49
     discriminated
    -1.49
    conservancy
    -1.46
     brittle
    -1.41
     proport
    -1.40
     roots
    -1.37
    quartered
    -1.32
     flattened
    -1.31
     uniformly
    -1.31
    POSITIVE LOGITS
    Guest
    1.65
    Reviewer
    1.41
    Cele
    1.36
    OPLE
    1.32
    Episode
    1.29
    Live
    1.29
    KING
    1.27
     Guest
    1.25
    rawdownloadcloneembedreportprint
    1.25
     Ask
    1.24
    Activations Density 0.007%

    No Known Activations