© 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. 13747
    Prev
    Next
    INDEX
    Explanations

    information related to sales promotions and availability of products, including discounts and availability for specific devices

    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
    579
    +0.12
    0.6%
    477
    +0.10
    0.4%
    629
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    579
    +0.12
    0.03
    609
    +0.10
    0.03
    245
    +0.10
    0.03
    Negative Logits
    rious
    -0.78
    orman
    -0.76
    orio
    -0.71
    rums
    -0.70
    stem
    -0.68
     overshadow
    -0.67
    lam
    -0.66
    ppel
    -0.66
    ron
    -0.65
    acter
    -0.64
    POSITIVE LOGITS
     download
    0.94
     downloadable
    0.90
     abundance
    0.82
     packages
    0.79
     downloads
    0.77
    taboola
    0.77
     subscription
    0.75
     ZIP
    0.75
    download
    0.72
    packages
    0.71
    Activations Density 0.156%

    No Known Activations