© 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. 12060
    Prev
    Next
    INDEX
    Explanations

    instances of text relating to political and social issues, including controversies, discrimination, scientific debates, and economic challenges

    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
    314
    +0.12
    0.5%
    199
    +0.09
    0.4%
    253
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    253
    +0.12
    0.09
    199
    +0.09
    0.09
    489
    +0.09
    0.09
    Negative Logits
     Bris
    -0.71
     Sapphire
    -0.69
     guiActiveUnfocused
    -0.68
     indo
    -0.65
     Bengal
    -0.64
    iewicz
    -0.64
    creen
    -0.63
     confines
    -0.63
     detached
    -0.63
     Opera
    -0.63
    POSITIVE LOGITS
    IJ
    1.13
    ª
    1.12
    ¹
    1.12
    ł
    1.09
    Ĵ
    1.08
    ı
    1.07
    ij
    1.03
    £
    0.94
    ³
    0.93
    certain
    0.91
    Activations Density 0.144%

    No Known Activations