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 49152
    5. 8-RES_FS49152-JB
    6. 26292
    Prev
    Next
    INDEX
    Explanations

    terms related to financial transactions and repayments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    318
    +0.10
    0.5%
    59
    +0.09
    0.4%
    532
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    245
    +0.10
    0.01
    318
    +0.09
    0.01
    405
    +0.09
    0.01
    Negative Logits
    quickShipAvailable
    -0.79
    oute
    -0.74
    ĸļ
    -0.73
    wcs
    -0.72
    atus
    -0.70
    ãĤ¦ãĤ¹
    -0.70
    antry
    -0.70
    aha
    -0.69
     differe
    -0.69
    iqueness
    -0.68
    POSITIVE LOGITS
     debts
    1.45
     debt
    1.29
     loans
    1.19
     creditors
    1.14
     dues
    1.14
     owed
    1.07
     Debt
    1.05
     mortgages
    1.00
     bills
    0.99
     loan
    0.97
    Activations Density 0.020%

    No Known Activations