© 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. 5-ATT-KK
    11. 24722
    Prev
    Next
    INDEX
    Explanations

    phrases that refer to the concept of parts or components in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.11
    1:0.12
    2:0.04
    3:0.05
    4:0.05
    5:0.22
    6:0.03
    7:0.04
    8:0.11
    9:0.06
    10:0.07
    11:0.07
    Negative Logits
     Mouth
    -1.55
     "%
    -1.52
     coefficients
    -1.49
     Rankings
    -1.45
    idth
    -1.43
     Spoon
    -1.35
    onic
    -1.34
     Records
    -1.34
     1024
    -1.31
    onics
    -1.31
    POSITIVE LOGITS
    etheless
    1.78
    stay
    1.75
    asio
    1.55
    SPONSORED
    1.44
    aith
    1.42
    speak
    1.40
    sic
    1.39
    soc
    1.38
    working
    1.38
    morrow
    1.38
    Activations Density 0.003%

    No Known Activations