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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 20555
    Prev
    Next
    INDEX
    Explanations

    numerical references and statistics related to injury and fatality counts in news stories

    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
    0:0.03
    1:0.02
    2:0.09
    3:0.21
    4:0.02
    5:0.07
    6:0.02
    7:0.08
    8:0.02
    9:0.02
    10:0.34
    11:0.03
    Negative Logits
     bandwagon
    -2.71
     wisely
    -2.32
     ideas
    -2.30
     dates
    -2.28
    eday
    -2.27
     hints
    -2.26
     formulas
    -2.24
     confidently
    -2.23
     nowadays
    -2.21
     Practices
    -2.19
    POSITIVE LOGITS
     injuring
    2.94
     amput
    2.81
    Damage
    2.77
     damages
    2.75
     injure
    2.66
     incapac
    2.65
     wounding
    2.64
     loss
    2.56
     damage
    2.55
     injured
    2.49
    Activations Density 0.150%

    No Known Activations