© 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. 12483
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate the initiation of various types of groups or relationships

    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
    0:0.04
    1:0.03
    2:0.02
    3:0.07
    4:0.05
    5:0.05
    6:0.05
    7:0.04
    8:0.02
    9:0.09
    10:0.03
    11:0.45
    Negative Logits
     proportions
    -2.44
     figures
    -2.42
     slips
    -2.35
     Figures
    -2.31
     weaknesses
    -2.29
    nature
    -2.22
    etheless
    -2.19
     strengths
    -2.17
     foresee
    -2.16
     quantities
    -2.15
    POSITIVE LOGITS
     thread
    2.71
     hobby
    2.70
     conversation
    2.67
     subreddit
    2.63
     discussion
    2.48
     feud
    2.44
     Discussion
    2.44
     lineage
    2.38
    ongo
    2.34
    thread
    2.27
    Activations Density 0.053%

    No Known Activations