© 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 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. 10824
    Prev
    Next
    INDEX
    Explanations

    keywords related to third-party components or services in software development

    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
    568
    +0.15
    0.7%
    763
    +0.10
    0.5%
    205
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    568
    +0.15
    0.02
    656
    +0.10
    0.01
    205
    +0.10
    0.01
    Negative Logits
    DAY
    -0.98
    frey
    -0.79
    zza
    -0.77
    Quantity
    -0.74
    bats
    -0.74
    Hour
    -0.74
    terday
    -0.73
    utenberg
    -0.73
    aturday
    -0.71
    speed
    -0.71
    POSITIVE LOGITS
     vendors
    0.97
     validation
    0.91
     sellers
    0.89
     intermediary
    0.87
     vendor
    0.87
     endorsement
    0.83
     endorsements
    0.83
     solutions
    0.82
     advertis
    0.79
     advertisers
    0.79
    Activations Density 0.032%

    No Known Activations