© 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. 9431
    Prev
    Next
    INDEX
    Explanations

    names of political figures and government officials

    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
    36
    +0.10
    0.5%
    521
    +0.10
    0.5%
    238
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.10
    232
    +0.10
    0.08
    521
    +0.10
    0.04
    Negative Logits
     robbers
    -0.72
    rawdownloadcloneembedreportprint
    -0.70
    CRIP
    -0.68
    acters
    -0.68
     phantom
    -0.65
    Stud
    -0.63
    ãĥ¼ãĤ¯
    -0.63
     skelet
    -0.63
     interpol
    -0.62
    rint
    -0.61
    POSITIVE LOGITS
     appoint
    0.88
    luck
    0.83
     appointed
    0.80
     Jinping
    0.78
     Hague
    0.75
     appointing
    0.75
     Minister
    0.72
    anyahu
    0.72
     Cabinet
    0.71
    gov
    0.71
    Activations Density 0.430%

    No Known Activations