© 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 12288
    9. 8-RES_FS12288-JB
    11. 10562
    Prev
    Next
    INDEX
    Explanations

    content related to political and social issues, including commentary on politicians and policies

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    +0.28
    1.4%
    143
    +0.10
    0.5%
    649
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.28
    0.18
    94
    +0.10
    0.14
    143
    +0.10
    0.10
    Negative Logits
    iton
    -0.75
    senal
    -0.74
     Reincarnated
    -0.69
    awei
    -0.65
    ktop
    -0.64
     Trog
    -0.63
    apon
    -0.62
    ction
    -0.61
    utic
    -0.61
     whereabouts
    -0.61
    POSITIVE LOGITS
    ³³³³³³³³
    1.11
    ³³³³
    1.08
    ³³³
    1.07
    ³³³³³³³³³³³³³³³³
    1.06
    Consider
    1.00
    Indeed
    0.93
    Moreover
    0.89
    Yet
    0.88
    Unless
    0.87
    Nevertheless
    0.87
    Activations Density 0.551%

    No Known Activations