© 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. 8031
    Prev
    Next
    INDEX
    Explanations

    the word "departure"

    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
    27
    +0.10
    0.5%
    288
    +0.10
    0.4%
    43
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    199
    +0.10
    0.02
    651
    +0.10
    0.01
    147
    +0.09
    0.01
    Negative Logits
    ibur
    -0.79
    verbs
    -0.73
    iop
    -0.72
    kered
    -0.69
    ebook
    -0.68
    flix
    -0.65
    ibu
    -0.65
    Cola
    -0.64
    essee
    -0.63
    ella
    -0.63
    POSITIVE LOGITS
     undone
    0.80
     from
    0.78
    ments
    0.69
     airport
    0.69
    doms
    0.66
     departure
    0.63
    ment
    0.62
     depart
    0.62
    untled
    0.61
     Wast
    0.61
    Activations Density 0.026%

    No Known Activations