© 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. 3-ATT-KK
    11. 13577
    Prev
    Next
    INDEX
    Explanations

    sentences expressing opinion or advocating for certain beliefs or actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    phrases that express positive attributes or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.04
    1:0.03
    2:0.09
    3:0.08
    4:0.06
    5:0.04
    6:0.11
    7:0.15
    8:0.11
    9:0.05
    10:0.11
    11:0.06
    Negative Logits
    omi
    -1.30
    ":"/
    -1.29
    atown
    -1.19
    urger
    -1.13
    rete
    -1.10
    mallow
    -1.09
    itudinal
    -1.08
    amera
    -1.07
    ropolis
    -1.07
    owe
    -1.06
    POSITIVE LOGITS
    seekers
    1.20
    ���
    1.11
     cues
    1.08
     infiltration
    1.08
     shading
    1.04
     Gleaming
    1.01
     juven
    0.99
     dwarves
    0.98
     conditioning
    0.98
     Journals
    0.97
    Activations Density 0.239%

    No Known Activations