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 24576
    5. 8-RES_FS24576-JB
    6. 23498
    Prev
    Next
    INDEX
    Explanations

    proposals for governmental and regulatory reforms regarding technology and societal issues

    oai_token-act-pair · gpt-4o-miniTriggered 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
    481
    +0.25
    1.2%
    447
    +0.15
    0.7%
    275
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    529
    +0.25
    0.06
    665
    +0.15
    0.08
    481
    +0.10
    0.10
    Negative Logits
    inventoryQuantity
    -0.86
     wounds
    -0.76
     Errors
    -0.74
     partName
    -0.70
    mist
    -0.64
    OGR
    -0.64
    stress
    -0.64
     sorrow
    -0.63
     disappoint
    -0.62
     wounding
    -0.61
    POSITIVE LOGITS
     someday
    1.20
     decentral
    1.00
     privat
    0.92
     decriminal
    0.91
     legalized
    0.91
     2020
    0.90
     interoper
    0.89
     abol
    0.89
     2030
    0.89
     2025
    0.89
    Activations Density 1.399%

    No Known Activations