© 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. 11863
    Prev
    Next
    INDEX
    Explanations

    vision-related words and phrases

    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
    475
    +0.12
    0.5%
    597
    +0.10
    0.4%
    762
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    435
    +0.12
    0.03
    505
    +0.10
    0.03
    475
    +0.10
    0.03
    Negative Logits
    emetery
    -0.69
     OPEC
    -0.69
     Slaughter
    -0.69
    atern
    -0.69
     Git
    -0.68
    lease
    -0.67
    sale
    -0.66
    cca
    -0.66
    aternal
    -0.64
     incentiv
    -0.64
    POSITIVE LOGITS
     impaired
    1.14
     goggles
    1.09
    ight
    1.04
     impairment
    1.01
     illusions
    0.94
     cones
    0.90
     comprehension
    0.87
    blind
    0.87
     blurry
    0.87
    ocular
    0.87
    Activations Density 0.137%

    No Known Activations