© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 2-ATT-KK
    11. 23273
    Prev
    Next
    INDEX
    Explanations

    commands or instructions in a technical context

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

    instances of output or results in technical contexts

    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
    0:0.04
    1:0.04
    2:0.19
    3:0.06
    4:0.08
    5:0.07
    6:0.03
    7:0.04
    8:0.09
    9:0.20
    10:0.06
    11:0.03
    Negative Logits
     holidays
    -1.27
    神
    -1.24
    Georg
    -1.23
    ayan
    -1.22
     christ
    -1.21
    istar
    -1.21
     dracon
    -1.20
    iba
    -1.17
     manic
    -1.16
    olog
    -1.15
    POSITIVE LOGITS
    output
    1.48
     outputs
    1.39
    intensive
    1.37
    Output
    1.27
    producing
    1.26
     Output
    1.25
     output
    1.24
     Forth
    1.23
    ngth
    1.23
    blers
    1.22
    Activations Density 0.008%

    No Known Activations