© 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. 13453
    Prev
    Next
    INDEX
    Explanations

    texts in different languages

    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
    385
    +0.13
    0.6%
    284
    +0.11
    0.5%
    616
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.01
    107
    +0.11
    0.01
    462
    +0.09
    0.01
    Negative Logits
    lisher
    -0.75
     streak
    -0.70
    teen
    -0.66
     bidder
    -0.65
    rences
    -0.65
    ters
    -0.65
     stag
    -0.63
     ribbon
    -0.63
     crossover
    -0.62
     flo
    -0.61
    POSITIVE LOGITS
    ãģĨ
    1.06
    ا
    0.96
    е
    0.91
    å®
    0.90
    ÑĮ
    0.89
    и
    0.87
    ÙĪ
    0.85
    Ãł
    0.85
    à¤
    0.82
    ocide
    0.82
    Activations Density 0.012%

    No Known Activations