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. 22392
    Prev
    Next
    INDEX
    Explanations

    political terms and references, especially related to policy and legislation

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

    names of prominent political figures, particularly those involved in the Republican Party

    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
    0:0.05
    1:0.04
    2:0.12
    3:0.07
    4:0.22
    5:0.06
    6:0.03
    7:0.05
    8:0.08
    9:0.09
    10:0.10
    11:0.06
    Negative Logits
    oglu
    -1.59
     TF
    -1.33
     largeDownload
    -1.32
    ptin
    -1.30
    utical
    -1.29
    ndra
    -1.27
    tics
    -1.25
     fest
    -1.25
    ı
    -1.25
     furnished
    -1.19
    POSITIVE LOGITS
    utherford
    1.53
     Majority
    1.43
    aud
    1.38
     senate
    1.37
     integrity
    1.33
     Choice
    1.29
     colleg
    1.28
     seiz
    1.28
     thous
    1.26
    ributed
    1.26
    Activations Density 0.019%

    No Known Activations