Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 5-ATT-KK
    6. 2874
    Prev
    Next
    INDEX
    Explanations

    instances of regret and remorse

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.02
    1:0.02
    2:0.04
    3:0.07
    4:0.12
    5:0.02
    6:0.03
    7:0.41
    8:0.04
    9:0.03
    10:0.09
    11:0.06
    Negative Logits
    ibaba
    -1.85
    inals
    -1.79
    lite
    -1.71
    ulhu
    -1.70
    riott
    -1.68
    soDeliveryDate
    -1.58
    izen
    -1.57
    resistant
    -1.55
    phabet
    -1.54
    izontal
    -1.53
    POSITIVE LOGITS
     incurred
    1.91
     mishand
    1.80
     amput
    1.78
     postp
    1.76
     abandonment
    1.75
     relinqu
    1.74
     recalling
    1.72
     forg
    1.71
     forfe
    1.70
     betrayal
    1.68
    Activations Density 0.007%

    No Known Activations