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. 10-ATT-KK
    6. 14607
    Prev
    Next
    INDEX
    Explanations

    sentences that conclude with strong punctuation or statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.12
    1:0.11
    2:0.09
    3:0.07
    4:0.06
    5:0.05
    6:0.10
    7:0.07
    8:0.05
    9:0.05
    10:0.13
    11:0.06
    Negative Logits
    MpServer
    -3.65
     Ur
    -3.27
     Harbaugh
    -3.25
     Arpaio
    -3.07
     Aval
    -3.05
    baugh
    -3.03
     XY
    -3.00
    Seg
    -2.99
    王
    -2.97
    ayers
    -2.88
    POSITIVE LOGITS
     Natalie
    6.30
    atalie
    4.30
     Judd
    3.60
    nit
    3.51
    Nit
    3.24
     NAS
    3.15
     Nigel
    2.97
     Swan
    2.81
    reens
    2.77
     Nom
    2.77
    Activations Density 0.000%

    No Known Activations