© 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. 24504
    Prev
    Next
    INDEX
    Explanations

    specific makes and models of vehicles and performance specifications

    oai_token-act-pair · gpt-4o-miniTriggered 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
    355
    +0.09
    0.4%
    664
    +0.09
    0.4%
    433
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.09
    0.05
    343
    +0.09
    0.05
    99
    +0.09
    0.04
    Negative Logits
    iquette
    -0.82
     Osw
    -0.78
    opian
    -0.70
    blogs
    -0.70
     oneself
    -0.67
    #$#$
    -0.67
    acio
    -0.67
    ocating
    -0.66
    ocrates
    -0.65
     lik
    -0.65
    POSITIVE LOGITS
     Specifications
    1.01
     Adapter
    0.94
     Dual
    0.91
     Hybrid
    0.88
     chassis
    0.87
     Titanium
    0.86
     Edition
    0.86
     Aluminum
    0.86
     Deluxe
    0.86
     variant
    0.85
    Activations Density 0.316%

    No Known Activations