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

    Neuronpedia

    Jacobian LensNEW
    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. 18023
    Prev
    Next
    INDEX
    Explanations

    information related to specific events or occurrences, such as winners of elections or closures of businesses

    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
    126
    +0.11
    0.6%
    28
    +0.08
    0.4%
    314
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.11
    0.08
    264
    +0.08
    0.02
    362
    +0.08
    0.05
    Negative Logits
     billions
    -0.72
     inherently
    -0.72
     ingrained
    -0.66
     unparalleled
    -0.65
     trillions
    -0.65
     fundamentally
    -0.64
     millions
    -0.63
     fueling
    -0.61
     ballpark
    -0.61
     capitalism
    -0.59
    POSITIVE LOGITS
    qus
    0.77
    çĶ°
    0.74
    uese
    0.71
    Others
    0.70
    onyms
    0.70
    ipal
    0.69
    éĥ
    0.68
    ļéĨĴ
    0.68
    irin
    0.68
     latter
    0.67
    Activations Density 0.877%

    No Known Activations