© 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 49152
    9. 8-RES_FS49152-JB
    11. 41639
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to open source software

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    668
    +0.10
    0.5%
    115
    +0.10
    0.5%
    362
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    668
    +0.10
    0.01
    115
    +0.10
    0.01
    188
    +0.09
    0.01
    Negative Logits
    ellen
    -0.88
    awed
    -0.83
    ques
    -0.79
    asin
    -0.77
    oola
    -0.75
    igne
    -0.74
    ousy
    -0.74
    awks
    -0.73
    _-
    -0.73
    iaz
    -0.72
    POSITIVE LOGITS
     software
    1.02
     repositories
    1.00
     repository
    0.94
     projects
    0.93
     project
    0.86
     platform
    0.85
     technologies
    0.85
     tools
    0.83
     platforms
    0.83
     alternatives
    0.82
    Activations Density 0.012%

    No Known Activations