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 1536
    5. 8-RES_FS1536-JB
    6. 817
    Prev
    Next
    INDEX
    Explanations

    quotations within the text with potential emotional or controversial impact

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    314
    +0.23
    1.1%
    102
    +0.20
    1.0%
    442
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.23
    0.13
    160
    +0.20
    0.11
    442
    +0.15
    0.10
    Negative Logits
     affiliate
    -0.79
     Ross
    -0.77
     rental
    -0.75
     outfielder
    -0.73
     matter
    -0.73
     rall
    -0.72
     grasp
    -0.72
     editor
    -0.71
     buoy
    -0.71
     McKay
    -0.70
    POSITIVE LOGITS
    classic
    1.45
    false
    1.44
    true
    1.44
    pure
    1.43
    little
    1.40
    normal
    1.40
    every
    1.38
    almost
    1.37
    double
    1.37
    nothing
    1.35
    Activations Density 0.524%

    No Known Activations