© 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 EvalsExportsinterp-engineNEWAPI 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. 16734
    Prev
    Next
    INDEX
    Explanations

    personal stories or anecdotes

    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
    640
    +0.30
    1.5%
    107
    +0.15
    0.7%
    160
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.30
    0.14
    107
    +0.15
    0.13
    160
    +0.14
    0.13
    Negative Logits
     bloc
    -0.83
     negoti
    -0.79
     displacement
    -0.79
     orchestr
    -0.78
     undet
    -0.77
     uranium
    -0.77
     undisclosed
    -0.76
     payroll
    -0.76
     antit
    -0.76
     disputed
    -0.76
    POSITIVE LOGITS
    Anyway
    1.90
    Advertisements
    1.56
    So
    1.52
    Advertisement
    1.50
    My
    1.48
    Thank
    1.48
    Because
    1.47
    Then
    1.46
    Being
    1.46
    However
    1.46
    Activations Density 0.390%

    No Known Activations