© 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. 44015
    Prev
    Next
    INDEX
    Explanations

    requests for audience feedback and thoughts

    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
    0:0.03
    1:0.01
    2:0.09
    3:0.11
    4:0.14
    5:0.03
    6:0.06
    7:0.22
    8:0.04
    9:0.05
    10:0.06
    11:0.10
    Negative Logits
     faked
    -1.24
    ties
    -1.23
     suits
    -1.22
    pred
    -1.18
     whiff
    -1.17
     dinosaurs
    -1.15
     videot
    -1.14
     vanished
    -1.12
     satellites
    -1.12
    ��極
    -1.12
    POSITIVE LOGITS
     clarification
    1.45
     educate
    1.42
     inquire
    1.42
     accordingly
    1.41
    zai
    1.40
     cautiously
    1.40
    arse
    1.39
     opin
    1.38
     someday
    1.38
     enthusi
    1.38
    Activations Density 0.004%

    No Known Activations