© 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
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 18992
    Prev
    Next
    INDEX
    Explanations

    the presence of verbs and actions associated with subjects

    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
    0:0.04
    1:0.02
    2:0.04
    3:0.16
    4:0.02
    5:0.05
    6:0.01
    7:0.04
    8:0.02
    9:0.01
    10:0.52
    11:0.01
    Negative Logits
     hates
    -2.30
     Wrong
    -2.24
     hurts
    -2.14
     Concern
    -2.08
     regretted
    -2.01
    threat
    -1.97
    wrong
    -1.96
     hurting
    -1.90
     lately
    -1.89
     Danger
    -1.88
    POSITIVE LOGITS
     access
    2.77
     unrestricted
    2.67
     uninterrupted
    2.54
     easier
    2.52
     unlimited
    2.52
     flexibility
    2.50
     finer
    2.46
    神
    2.42
    �
    2.40
     bypass
    2.31
    Activations Density 0.231%

    No Known Activations