© 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. 10432
    Prev
    Next
    INDEX
    Explanations

    contact information and promotional text resembling advertisements

    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
    87
    +0.27
    1.5%
    640
    +0.26
    1.5%
    481
    +0.19
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.27
    0.13
    160
    +0.26
    0.13
    442
    +0.19
    0.11
    Negative Logits
     increasingly
    -0.79
     spir
    -0.78
     brid
    -0.75
     sculpt
    -0.74
     perce
    -0.74
     choking
    -0.73
     loos
    -0.73
     reneg
    -0.72
     quir
    -0.72
     labor
    -0.71
    POSITIVE LOGITS
    Advertisements
    1.56
    Conclusion
    1.50
    Discussion
    1.35
    References
    1.34
    Translation
    1.28
    Loading
    1.25
    Official
    1.25
    --------------------------------
    1.24
    ================================================================
    1.24
    Send
    1.23
    Activations Density 0.335%

    No Known Activations