© 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. 7755
    Prev
    Next
    INDEX
    Explanations

    terms related to software performance improvements

    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
    288
    +0.14
    0.6%
    724
    +0.10
    0.5%
    216
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    724
    +0.14
    0.02
    540
    +0.10
    0.02
    288
    +0.09
    0.02
    Negative Logits
    sonian
    -0.92
    shi
    -0.82
    ril
    -0.82
    idan
    -0.76
    yu
    -0.75
    raine
    -0.74
     Passage
    -0.74
    eton
    -0.72
     Voices
    -0.71
    zona
    -0.69
    POSITIVE LOGITS
     optimizations
    1.22
     optimization
    1.18
     improvements
    1.06
     enhancement
    1.03
     improvement
    1.03
     enhancing
    1.03
     enhancements
    1.02
     degradation
    1.02
     regress
    0.99
     benchmarks
    0.99
    Activations Density 0.031%

    No Known Activations