Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 5-ATT-KK
    6. 43841
    Prev
    Next
    INDEX
    Explanations

    phrases and words indicating roles, positions, and actions related to work and responsibility

    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
    0:0.01
    1:0.01
    2:0.12
    3:0.05
    4:0.22
    5:0.02
    6:0.03
    7:0.17
    8:0.03
    9:0.03
    10:0.12
    11:0.13
    Negative Logits
    mma
    -1.36
    doms
    -1.35
    Winged
    -1.34
    istine
    -1.32
    pir
    -1.29
    OTOS
    -1.26
    llah
    -1.23
    yg
    -1.22
    Reviewer
    -1.22
    icol
    -1.22
    POSITIVE LOGITS
     Vec
    1.38
     upkeep
    1.21
     Kend
    1.16
     Falk
    1.13
     Burnett
    1.11
     Uri
    1.11
     Decay
    1.11
     Reach
    1.09
     HS
    1.07
     EB
    1.06
    Activations Density 0.122%

    No Known Activations