Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 776
    Prev
    Next
    INDEX
    Explanations

    banking and financial terms and entities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    120
    +0.10
    0.4%
    35
    +0.10
    0.4%
    270
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.15
    35
    +0.10
    0.07
    592
    +0.09
    0.06
    Negative Logits
     Ridley
    -0.72
     Duncan
    -0.72
    IRO
    -0.69
    SPA
    -0.68
    advertising
    -0.67
    çĦ
    -0.66
    wic
    -0.66
    ILE
    -0.65
     Vu
    -0.65
    gha
    -0.65
    POSITIVE LOGITS
     liquidity
    1.14
     repayment
    1.10
     transaction
    1.10
     lenders
    1.10
     transactions
    1.08
    urrencies
    1.08
     banking
    1.07
     issuer
    1.05
     lender
    1.04
     bankers
    1.03
    Activations Density 2.317%

    No Known Activations