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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 12680
    Prev
    Next
    INDEX
    Explanations

    references to agreements or deals in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z
    Hook Layer
    9
    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
    1:0.02
    2:0.40
    3:0.09
    4:0.03
    5:0.08
    6:0.02
    7:0.05
    8:0.06
    9:0.03
    10:0.03
    11:0.01
    Negative Logits
    Redditor
    -2.85
    opath
    -2.75
     surn
    -2.60
    Origin
    -2.57
     Insp
    -2.57
    ilial
    -2.41
    infect
    -2.37
    diagn
    -2.33
     Ort
    -2.32
    esity
    -2.32
    POSITIVE LOGITS
     negotiators
    6.86
     negotiations
    6.19
     negotiating
    6.11
     negotiation
    6.05
     negotiator
    5.89
     negotiated
    5.88
     renegoti
    5.26
     agreement
    5.21
     agreements
    5.20
     bargaining
    5.09
    Activations Density 1.022%

    No Known Activations