© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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 3072
    9. 8-RES_FS3072-JB
    11. 535
    Prev
    Next
    INDEX
    Explanations

    research findings, studies, and reports on various topics, particularly related to work, education, and public services

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    447
    +0.14
    0.6%
    481
    +0.13
    0.6%
    375
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.14
    0.08
    576
    +0.13
    0.06
    142
    +0.12
    0.08
    Negative Logits
    alogue
    -0.76
    bury
    -0.71
    ãĥĩ
    -0.69
    Film
    -0.66
    mare
    -0.65
    gur
    -0.65
    wikipedia
    -0.64
    xtap
    -0.63
    ico
    -0.63
    ãĥ¬
    -0.63
    POSITIVE LOGITS
     widespread
    0.83
     disparities
    0.83
     correlations
    0.80
     discrepancies
    0.79
     significant
    0.79
     median
    0.79
     anomalies
    0.78
     conclusive
    0.78
     irregularities
    0.77
     probable
    0.77
    Activations Density 1.158%

    No Known Activations