© 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. 17263
    Prev
    Next
    INDEX
    Explanations

    structured lists or bullet points

    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
    640
    +0.30
    1.5%
    160
    +0.17
    0.9%
    442
    +0.16
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.30
    0.06
    160
    +0.17
    0.06
    442
    +0.16
    0.06
    Negative Logits
     retard
    -0.81
     revers
    -0.76
     scrap
    -0.76
     brid
    -0.76
     treaty
    -0.74
     consulate
    -0.73
     frontline
    -0.73
     regener
    -0.73
     masked
    -0.71
     negoti
    -0.70
    POSITIVE LOGITS
    Conclusion
    1.63
    Advertisements
    1.59
    CONCLUS
    1.46
    --------------------------------------------------------
    1.45
    Finally
    1.36
    ================================================================
    1.36
    Lastly
    1.32
    Contribut
    1.29
    References
    1.29
    ________________________________________________________________
    1.28
    Activations Density 0.114%

    No Known Activations