Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 26023
    Prev
    Next
    INDEX
    Explanations

    events or actions related to education or schooling

    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
    246
    +0.11
    0.5%
    445
    +0.11
    0.5%
    555
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.11
    0.02
    555
    +0.11
    0.02
    469
    +0.09
    0.01
    Negative Logits
    slice
    -0.78
    deal
    -0.75
    vich
    -0.73
    pex
    -0.70
    timer
    -0.70
    hl
    -0.69
    selves
    -0.69
    berman
    -0.69
    worn
    -0.69
    heading
    -0.68
    POSITIVE LOGITS
     university
    1.10
     college
    1.07
     school
    1.06
     colleges
    1.00
     graduate
    0.95
     Hogwarts
    0.90
     School
    0.89
     universities
    0.89
     Harvard
    0.88
     kindergarten
    0.88
    Activations Density 0.009%

    No Known Activations