© 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. 5762
    Prev
    Next
    INDEX
    Explanations

    sentences talking about historical events, judgments, and indicators

    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
    113
    +0.10
    0.4%
    633
    +0.09
    0.4%
    474
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    650
    +0.10
    0.05
    102
    +0.09
    0.05
    7
    +0.09
    0.04
    Negative Logits
    SpaceEngineers
    -0.79
    awaru
    -0.72
    ortment
    -0.69
    artney
    -0.69
    enegger
    -0.68
    successful
    -0.66
    thinkable
    -0.66
     undet
    -0.64
    iscopal
    -0.64
    æĸ¹
    -0.63
    POSITIVE LOGITS
     abound
    0.76
    Ĭ±
    0.70
     sciences
    0.67
     dictates
    0.67
     Thrones
    0.67
     dictate
    0.66
    goo
    0.66
     spilled
    0.66
     Hers
    0.63
     arbit
    0.61
    Activations Density 0.299%

    No Known Activations