© 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. 33228
    Prev
    Next
    INDEX
    Explanations

    tips and guidelines for effective practices, particularly focused on advice for newcomers or specific tasks

    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
    0:0.01
    1:0.01
    2:0.06
    3:0.04
    4:0.07
    5:0.02
    6:0.05
    7:0.43
    8:0.02
    9:0.02
    10:0.14
    11:0.08
    Negative Logits
    SpaceEngineers
    -1.55
    apsed
    -1.53
    lishes
    -1.45
     leased
    -1.44
    actionDate
    -1.44
     partName
    -1.42
     respectively
    -1.40
    76561
    -1.39
     occupancy
    -1.38
    data
    -1.36
    POSITIVE LOGITS
     pitfalls
    1.67
     Helpful
    1.59
     Tackle
    1.57
     stocking
    1.54
     Sneak
    1.54
    Avoid
    1.53
    ��
    1.50
     Avoid
    1.49
     hygiene
    1.48
     trick
    1.46
    Activations Density 0.009%

    No Known Activations