© 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. 22399
    Prev
    Next
    INDEX
    Explanations

    references to academic studies or research findings

    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
    575
    +0.12
    0.5%
    756
    +0.10
    0.5%
    651
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    574
    +0.12
    0.03
    419
    +0.10
    0.03
    651
    +0.10
    0.03
    Negative Logits
    rising
    -0.63
    ansas
    -0.61
    fw
    -0.60
    oji
    -0.59
     Fiesta
    -0.58
     overloaded
    -0.57
     fitt
    -0.57
     Reyn
    -0.56
     defective
    -0.56
     notor
    -0.56
    POSITIVE LOGITS
    udo
    0.98
     conducted
    0.89
    study
    0.82
     abroad
    0.81
     Subjects
    0.79
    Study
    0.77
    sonian
    0.73
     methodology
    0.72
    iveness
    0.72
     Researchers
    0.72
    Activations Density 0.042%

    No Known Activations