© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 2-ATT-KK
    6. 23271
    Prev
    Next
    INDEX
    Explanations

    financial terms and statistics

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

    terms associated with the concept of yield and its implications in various contexts, such as benefits, information, and economic aspects

    oai_token-act-pair · gpt-4o-miniTriggered 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
    0:0.07
    1:0.03
    2:0.13
    3:0.10
    4:0.05
    5:0.08
    6:0.02
    7:0.02
    8:0.27
    9:0.11
    10:0.04
    11:0.02
    Negative Logits
    ��
    -1.48
     operated
    -1.22
    ▬
    -1.14
     discriminated
    -1.09
     wors
    -1.09
     metic
    -1.06
     rang
    -1.06
     Helena
    -1.06
     Leigh
    -1.05
     cared
    -1.03
    POSITIVE LOGITS
    bye
    1.29
    lethal
    1.26
    ptoms
    1.25
    >)
    1.24
    >[
    1.20
     immortality
    1.19
    grain
    1.18
     %%
    1.16
     SPACE
    1.16
     ire
    1.15
    Activations Density 0.009%

    No Known Activations