© 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. 20724
    Prev
    Next
    INDEX
    Explanations

    legal and judicial-related terms

    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
    297
    +0.10
    0.5%
    324
    +0.09
    0.4%
    591
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.10
    -0.01
    103
    +0.09
    0.03
    284
    +0.09
    0.03
    Negative Logits
    eele
    -0.73
     Falk
    -0.66
     Loss
    -0.66
     Liver
    -0.66
     Tuc
    -0.66
     Spa
    -0.65
     Brow
    -0.64
    borough
    -0.64
     Kev
    -0.64
     Requ
    -0.62
    POSITIVE LOGITS
    ante
    0.81
    APH
    0.81
    naissance
    0.80
    antes
    0.79
    antly
    0.77
    aded
    0.76
    ainment
    0.75
    oad
    0.75
    oned
    0.73
    ect
    0.72
    Activations Density 0.082%

    No Known Activations