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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9455
    Prev
    Next
    INDEX
    Explanations

    references to scientific research and government initiatives related to specific fields or programs

    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
    110
    +0.11
    0.5%
    232
    +0.10
    0.4%
    479
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    110
    +0.11
    0.06
    247
    +0.10
    0.05
    288
    +0.10
    0.05
    Negative Logits
    inventoryQuantity
    -0.90
     spoilers
    -0.79
    stellar
    -0.76
    quickShipAvailable
    -0.76
    yond
    -0.73
    luaj
    -0.71
     quartz
    -0.69
    magic
    -0.68
    paio
    -0.68
    humans
    -0.68
    POSITIVE LOGITS
     Association
    1.52
     Program
    1.40
     Initiative
    1.40
     Convention
    1.29
     Center
    1.28
     Fund
    1.28
     Complex
    1.27
     Act
    1.27
     Improvement
    1.26
     Laboratory
    1.25
    Activations Density 0.208%

    No Known Activations