© 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. 6680
    Prev
    Next
    INDEX
    Explanations

    information related to statistics, data analysis, and comparisons

    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
    640
    +0.14
    0.7%
    340
    +0.11
    0.5%
    293
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.14
    0.10
    149
    +0.11
    0.04
    6
    +0.10
    0.06
    Negative Logits
     interf
    -0.85
     evangel
    -0.81
     tant
    -0.81
     exorc
    -0.77
     lect
    -0.76
     prayers
    -0.73
     commun
    -0.72
     apolog
    -0.71
     pleas
    -0.70
     antit
    -0.70
    POSITIVE LOGITS
    verages
    1.55
     Rankings
    1.25
    Rank
    1.21
    adjusted
    1.21
     averages
    1.20
     Average
    1.19
    chart
    1.16
    Average
    1.15
     Adjusted
    1.13
     Median
    1.12
    Activations Density 0.624%

    No Known Activations