© 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. 14285
    Prev
    Next
    INDEX
    Explanations

    references to legal or judicial issues in educational settings

    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
    +0.20
    0.9%
    447
    +0.16
    0.7%
    104
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.20
    0.06
    247
    +0.16
    0.04
    455
    +0.10
    0.05
    Negative Logits
     Detected
    -0.77
     Helpful
    -0.76
    Ô
    -0.72
    NetMessage
    -0.71
    ynthesis
    -0.70
    aphael
    -0.70
    borgh
    -0.67
    thood
    -0.66
     Verse
    -0.66
    haar
    -0.66
    POSITIVE LOGITS
     administrator
    1.04
     owner
    1.01
     clerk
    0.99
     superintendent
    0.96
     inspector
    0.96
     chiefs
    0.92
     solicitor
    0.91
     manager
    0.90
     attorney
    0.89
     operator
    0.88
    Activations Density 0.249%

    No Known Activations