© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 1472
    Prev
    Next
    INDEX
    Explanations

    capitalized acronyms or abbreviations

    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
    373
    +0.34
    1.8%
    496
    +0.10
    0.5%
    121
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.34
    -0.01
    232
    +0.10
    0.05
    284
    +0.08
    0.04
    Negative Logits
     hurd
    -0.53
    enhagen
    -0.52
     Akin
    -0.52
     Faul
    -0.51
     Citation
    -0.51
     Ps
    -0.50
     Rowling
    -0.48
     Highlands
    -0.47
     tantal
    -0.47
     fixme
    -0.47
    POSITIVE LOGITS
    rage
    0.62
    aza
    0.62
    cot
    0.61
    henko
    0.60
    til
    0.60
    yk
    0.60
    ance
    0.59
    ania
    0.57
    oad
    0.56
    ross
    0.56
    Activations Density 0.169%

    No Known Activations