© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 10-ATT-KK
    11. 16834
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "on" and contexts involving legal or procedural actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z
    Hook Layer
    10
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.05
    1:0.36
    2:0.07
    3:0.05
    4:0.07
    5:0.03
    6:0.06
    7:0.06
    8:0.04
    9:0.04
    10:0.06
    11:0.05
    Negative Logits
     hotter
    -2.77
    iled
    -2.60
     hyper
    -2.57
     dispro
    -2.44
     chronic
    -2.41
     perpetual
    -2.39
     runaway
    -2.35
     ridiculed
    -2.35
     iron
    -2.32
     fetish
    -2.31
    POSITIVE LOGITS
     Good
    7.01
    Good
    6.25
    good
    4.76
     GOOD
    4.51
     Goodman
    3.95
     Fair
    3.59
     Goodwin
    3.35
     Excellent
    3.24
     Bad
    3.16
     Easy
    3.04
    Activations Density 0.024%

    No Known Activations