© 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. 3-ATT-KK
    11. 8678
    Prev
    Next
    INDEX
    Explanations

    instances of the word "like" used in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    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
    1:0.06
    2:0.17
    3:0.10
    4:0.02
    5:0.03
    6:0.08
    7:0.06
    8:0.09
    9:0.08
    10:0.08
    11:0.16
    Negative Logits
    Merit
    -1.60
    lees
    -1.43
     Ports
    -1.34
     Fathers
    -1.29
    uments
    -1.24
     Steele
    -1.23
     Farrell
    -1.23
    arers
    -1.23
    ahime
    -1.22
     Gle
    -1.21
    POSITIVE LOGITS
    cro
    1.34
     akin
    1.32
    worker
    1.31
    timer
    1.24
     extra
    1.23
     separates
    1.21
     awoken
    1.21
    caliber
    1.20
    imental
    1.19
     resembling
    1.18
    Activations Density 0.004%

    No Known Activations