© 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 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. 41806
    Prev
    Next
    INDEX
    Explanations

    phrases related to direct speech, particularly the word "said" and its variations

    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
    0:0.02
    1:0.02
    2:0.06
    3:0.08
    4:0.07
    5:0.12
    6:0.04
    7:0.04
    8:0.09
    9:0.12
    10:0.16
    11:0.13
    Negative Logits
    cens
    -1.56
    isSpecialOrderable
    -1.43
    ム
    -1.28
    ̶
    -1.27
    ipl
    -1.24
    gex
    -1.23
    bread
    -1.22
    olkien
    -1.22
    ッド
    -1.21
    akespe
    -1.19
    POSITIVE LOGITS
     contention
    1.27
    ',"
    1.22
     stunned
    1.21
     vehemently
    1.13
    cffff
    1.13
     falsely
    1.12
     concede
    1.10
     Joint
    1.09
     repeatedly
    1.09
     interrupted
    1.08
    Activations Density 0.101%

    No Known Activations