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. 2-ATT-KK
    6. 22479
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology, commentary on social issues, and personal narratives involving conversation and experience

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    phrases expressing uncertainty or questioning societal norms

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

    No Comments

    Head Attr Weights
    0:0.06
    1:0.05
    2:0.06
    3:0.07
    4:0.03
    5:0.13
    6:0.05
    7:0.09
    8:0.05
    9:0.10
    10:0.18
    11:0.09
    Negative Logits
    osa
    -0.93
    装
    -0.92
    iggins
    -0.90
    taboola
    -0.89
    acas
    -0.88
    acket
    -0.87
    gain
    -0.87
    iage
    -0.87
    knife
    -0.85
    bryce
    -0.85
    POSITIVE LOGITS
     spoilers
    1.11
     goddamn
    1.03
     ain
    1.01
     selves
    1.00
     Elven
    1.00
     apologies
    0.98
     maybe
    0.94
    !--
    0.92
     Allaah
    0.90
     🙂
    0.89
    Activations Density 0.474%

    No Known Activations