© 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 EvalsExportsAPI 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. 6098
    Prev
    Next
    INDEX
    Explanations

    misconduct and punishment-related terms and scenarios

    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
    749
    +0.10
    0.5%
    500
    +0.10
    0.5%
    581
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.10
    0.05
    8
    +0.10
    0.05
    566
    +0.10
    0.04
    Negative Logits
    knit
    -0.68
    ilogy
    -0.64
     Palmer
    -0.63
     Paddock
    -0.63
     Elliot
    -0.61
     Tess
    -0.60
    enture
    -0.60
     Normandy
    -0.60
     Sutherland
    -0.60
    ciples
    -0.59
    POSITIVE LOGITS
     dishon
    0.99
     unacceptable
    0.98
     unworthy
    0.90
    Failure
    0.87
     disqual
    0.87
    Invalid
    0.86
     disgrace
    0.86
     unfit
    0.85
     invalid
    0.84
     fail
    0.84
    Activations Density 0.523%

    No Known Activations