© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 2-ATT-KK
    11. 14636
    Prev
    Next
    INDEX
    Explanations

    small, vague, or simple everyday words and phrases

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    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
    2:0.25
    3:0.07
    4:0.05
    5:0.06
    6:0.02
    7:0.01
    8:0.32
    9:0.08
    10:0.03
    11:0.01
    Negative Logits
     arrivals
    -1.32
     unfocusedRange
    -1.15
     Chamber
    -1.15
    illary
    -1.15
    icka
    -1.14
    imester
    -1.11
    iott
    -1.10
    sted
    -1.09
     Classification
    -1.08
    represented
    -1.05
    POSITIVE LOGITS
    igham
    1.41
    chuk
    1.32
    puff
    1.26
    ieu
    1.26
     monkey
    1.22
    healthy
    1.21
     overboard
    1.21
    axter
    1.14
    atro
    1.12
    yp
    1.11
    Activations Density 0.078%

    No Known Activations