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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6742
    Prev
    Next
    INDEX
    Explanations

    phrases related to business operations and charitable activities

    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
    87
    +0.37
    2.0%
    459
    +0.17
    0.9%
    756
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.37
    -0.01
    459
    +0.17
    0.06
    756
    +0.12
    -0.02
    Negative Logits
    inventoryQuantity
    -0.64
     Errors
    -0.58
    wrong
    -0.57
     Disapp
    -0.53
     Tolkien
    -0.53
    errors
    -0.52
     erroneous
    -0.52
    entious
    -0.51
     wrongly
    -0.51
     flawed
    -0.51
    POSITIVE LOGITS
     preferably
    1.04
    soDeliveryDate
    0.97
     ideally
    0.84
     Ideally
    0.82
     will
    0.81
     feasibility
    0.80
     TBA
    0.80
     SHALL
    0.77
     2021
    0.77
     2020
    0.76
    Activations Density 1.010%

    No Known Activations