© 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. 8726
    Prev
    Next
    INDEX
    Explanations

    various video game titles and releases

    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
    0:0.02
    1:0.07
    2:0.02
    3:0.06
    4:0.10
    5:0.38
    6:0.02
    7:0.05
    8:0.05
    9:0.02
    10:0.03
    11:0.14
    Negative Logits
     Purg
    -2.34
     Blaz
    -2.33
     famine
    -2.20
     Grave
    -2.13
    urrection
    -2.11
     Coffin
    -2.10
     Hurricane
    -2.10
     haunt
    -2.08
     Resident
    -2.07
    essional
    -2.04
    POSITIVE LOGITS
    illance
    2.43
    �
    2.30
    �
    2.13
    attering
    2.08
     immature
    2.07
    ↵
    2.06
    udder
    2.02
    uan
    2.00
     wires
    1.99
     counterfe
    1.97
    Activations Density 0.087%

    No Known Activations