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. 3-ATT-KK
    6. 8593
    Prev
    Next
    INDEX
    Explanations

    references to statistical data and conditions

    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
    0:0.03
    1:0.09
    2:0.14
    3:0.04
    4:0.06
    5:0.08
    6:0.10
    7:0.11
    8:0.09
    9:0.05
    10:0.09
    11:0.06
    Negative Logits
    yip
    -1.25
     Strongh
    -1.23
    Haunted
    -1.07
    aucuses
    -0.99
     ADS
    -0.98
    andals
    -0.93
    iferation
    -0.93
    rossover
    -0.93
    idency
    -0.92
    abilia
    -0.92
    POSITIVE LOGITS
     ib
    1.22
     os
    1.13
     ■
    1.04
    hai
    1.01
     Indra
    0.96
     [(
    0.96
     prob
    0.95
    sing
    0.94
     thou
    0.93
     nob
    0.92
    Activations Density 0.068%

    No Known Activations