© 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. 2-ATT-KK
    11. 16004
    Prev
    Next
    INDEX
    Explanations

    transcripts from various conversations and sources

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Head Attr Weights
    0:0.05
    1:0.03
    2:0.17
    3:0.05
    4:0.16
    5:0.07
    6:0.03
    7:0.03
    8:0.06
    9:0.22
    10:0.05
    11:0.03
    Negative Logits
    ahime
    -1.47
    ubi
    -1.39
    wings
    -1.29
    iage
    -1.27
    rican
    -1.26
    ディ
    -1.25
    east
    -1.22
    akra
    -1.20
    athered
    -1.19
    uay
    -1.19
    POSITIVE LOGITS
     Transcript
    1.86
     transcripts
    1.74
     transcript
    1.73
    Quotes
    1.49
     PBS
    1.36
     Pastebin
    1.35
     transcription
    1.34
    opus
    1.29
     newsp
    1.25
     annotations
    1.25
    Activations Density 0.004%

    No Known Activations