© 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. 5-ATT-KK
    11. 44657
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly commas and colons, indicating structure and pauses in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    0:0.19
    1:0.14
    2:0.04
    3:0.05
    4:0.06
    5:0.11
    6:0.06
    7:0.03
    8:0.07
    9:0.06
    10:0.06
    11:0.08
    Negative Logits
     coh
    -1.87
    ynt
    -1.59
    assisted
    -1.51
    ynthesis
    -1.44
    atron
    -1.40
    覚醒
    -1.39
    uble
    -1.39
    zie
    -1.37
     coord
    -1.36
    inals
    -1.36
    POSITIVE LOGITS
    AMD
    1.40
    AME
    1.36
     '[
    1.35
     GDDR
    1.32
     embarrassing
    1.29
     '
    1.28
    ••
    1.26
     attRot
    1.23
     unbeliev
    1.23
    -|
    1.23
    Activations Density 0.001%

    No Known Activations