© 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. 5-ATT-KK
    11. 27163
    Prev
    Next
    INDEX
    Explanations

    frequent uses of the word "to" and other prepositions or conjunctions in a technical or detailed context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    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
    1:0.34
    2:0.03
    3:0.04
    4:0.03
    5:0.13
    6:0.03
    7:0.03
    8:0.06
    9:0.04
    10:0.04
    11:0.04
    Negative Logits
    ODY
    -2.00
    Problem
    -1.89
    Function
    -1.86
    ROM
    -1.78
    bery
    -1.71
    ibrary
    -1.71
    ignty
    -1.69
    iable
    -1.68
    mercial
    -1.68
    LU
    -1.67
    POSITIVE LOGITS
     add
    2.25
     introduce
    2.15
     Akin
    2.04
     appoint
    2.01
     erase
    2.00
     bring
    1.94
     remove
    1.93
     subtract
    1.93
     attach
    1.92
    Add
    1.91
    Activations Density 0.004%

    No Known Activations