© 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. 10-ATT-KK
    11. 20870
    Prev
    Next
    INDEX
    Explanations

    references to community and related sentiments

    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
    0:0.06
    1:0.07
    2:0.19
    3:0.07
    4:0.07
    5:0.07
    6:0.05
    7:0.05
    8:0.05
    9:0.08
    10:0.05
    11:0.13
    Negative Logits
    FW
    -2.77
     <-
    -2.75
     MIN
    -2.65
    ldom
    -2.61
    police
    -2.60
    NAS
    -2.56
     COURT
    -2.55
     SAR
    -2.54
    Cass
    -2.52
    DOM
    -2.50
    POSITIVE LOGITS
    plex
    2.71
     subclass
    2.66
     Escape
    2.59
     Dive
    2.56
     Abstract
    2.55
     Bore
    2.55
     Clair
    2.44
     Construct
    2.43
     Test
    2.42
     Feather
    2.41
    Activations Density 0.000%

    No Known Activations