© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 13523
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial and economic matters, particularly related to debt, payments, pensions, retirement, and banking systems

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.18
    0.9%
    732
    +0.12
    0.5%
    0
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    732
    +0.18
    0.04
    0
    +0.12
    0.05
    393
    +0.11
    0.06
    Negative Logits
    youtu
    -0.87
    heastern
    -0.75
    aclysm
    -0.73
     Trog
    -0.72
    oward
    -0.72
    ople
    -0.71
    heast
    -0.70
    achev
    -0.69
     rays
    -0.69
    ankind
    -0.69
    POSITIVE LOGITS
     entitle
    1.13
     citizenship
    1.04
     entitlement
    1.03
     mortgages
    1.02
     passports
    1.01
     dues
    1.00
     eligibility
    0.99
     confidentiality
    0.99
     vouchers
    0.98
     passport
    0.96
    Activations Density 0.304%

    No Known Activations