© 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. 7607
    Prev
    Next
    INDEX
    Explanations

    information about events, updates, and promotions

    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
    +0.17
    0.8%
    6
    +0.16
    0.7%
    218
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.17
    0.13
    640
    +0.16
    0.09
    145
    +0.09
    0.07
    Negative Logits
    ensibly
    -0.76
     reb
    -0.73
     inclusion
    -0.72
     acute
    -0.72
     differential
    -0.71
     nonexistent
    -0.71
     interpre
    -0.71
     precip
    -0.70
     infiltration
    -0.69
     reversed
    -0.69
    POSITIVE LOGITS
    <|endoftext|>
    1.57
     Hopefully
    1.27
     Amen
    1.22
    ________________________________
    1.15
    ********************************
    1.14
     Until
    1.13
     ;)
    1.12
     Enjoy
    1.11
     ðŁĻĤ
    1.11
    Advertisements
    1.10
    Activations Density 0.427%

    No Known Activations