© 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. 2-ATT-KK
    11. 22651
    Prev
    Next
    INDEX
    Explanations

    information related to academic studies and scholarly research

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    elements related to structure and consistency in argumentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    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
    1:0.02
    2:0.22
    3:0.05
    4:0.25
    5:0.05
    6:0.02
    7:0.03
    8:0.09
    9:0.11
    10:0.04
    11:0.02
    Negative Logits
    ITNESS
    -1.33
     spouse
    -1.26
    ilet
    -1.25
     Saber
    -1.23
     Anger
    -1.14
     spouses
    -1.12
     Aad
    -1.11
     newsp
    -1.11
     McGr
    -1.10
    ailability
    -1.08
    POSITIVE LOGITS
    estern
    1.49
    ipel
    1.46
    ⓘ
    1.29
    andom
    1.28
    Case
    1.26
    ento
    1.25
    orously
    1.22
    udo
    1.20
    asonic
    1.18
    rices
    1.16
    Activations Density 0.007%

    No Known Activations