© 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. 15908
    Prev
    Next
    INDEX
    Explanations

    information related to legal actions, bans, and legislation

    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
    481
    +0.22
    1.1%
    712
    +0.13
    0.6%
    53
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    712
    +0.22
    0.04
    665
    +0.13
    0.07
    608
    +0.10
    0.08
    Negative Logits
    Plot
    -0.69
    {\
    -0.69
    offer
    -0.68
     consisted
    -0.65
    inar
    -0.64
     Emin
    -0.63
     Orchestra
    -0.62
    DragonMagazine
    -0.62
    ³³³³³³³³³³³³³³³³
    -0.61
    ointment
    -0.61
    POSITIVE LOGITS
     policing
    1.02
     unregulated
    0.97
     taxpayer
    0.95
     accountability
    0.95
     legalized
    0.92
     discriminatory
    0.91
     unchecked
    0.89
     federally
    0.89
     invasive
    0.85
     unsafe
    0.84
    Activations Density 0.846%

    No Known Activations