© 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 EvalsExportsinterp-engineNEWAPI 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. 21154
    Prev
    Next
    INDEX
    Explanations

    references to Emory University

    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
    359
    +0.10
    0.5%
    730
    +0.10
    0.4%
    53
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    359
    +0.10
    0.04
    53
    +0.10
    0.04
    117
    +0.10
    0.04
    Negative Logits
    worthiness
    -0.77
     wagon
    -0.70
     hoard
    -0.64
    meaning
    -0.63
     corrid
    -0.62
     awoken
    -0.62
     filler
    -0.61
    arding
    -0.61
     domestically
    -0.60
    ichick
    -0.60
    POSITIVE LOGITS
     University
    1.69
    University
    1.43
     faculty
    1.32
     Faculty
    1.30
     professors
    1.21
     undergrad
    1.20
     universities
    1.19
     Universities
    1.18
     Students
    1.17
     students
    1.14
    Activations Density 0.097%

    No Known Activations