© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 5-ATT-KK
    6. 19622
    Prev
    Next
    INDEX
    Explanations

    phrases related to measurements or quantities, particularly in culinary or food contexts

    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
    0:0.04
    1:0.02
    2:0.07
    3:0.06
    4:0.06
    5:0.03
    6:0.12
    7:0.28
    8:0.04
    9:0.03
    10:0.13
    11:0.07
    Negative Logits
     leaps
    -1.42
     concluding
    -1.41
     Suc
    -1.41
     dramatic
    -1.39
     discovering
    -1.37
     diagn
    -1.36
     intuitive
    -1.36
     investigation
    -1.35
    hooting
    -1.34
     definitively
    -1.34
    POSITIVE LOGITS
    plates
    1.87
     plates
    1.62
    Glass
    1.59
    isites
    1.53
     bowls
    1.52
    expensive
    1.49
    bags
    1.49
     plate
    1.49
    ividually
    1.49
    cheon
    1.47
    Activations Density 0.001%

    No Known Activations