© 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. 22009
    Prev
    Next
    INDEX
    Explanations

    words related to investigations, specifically the mention of the special counsel Robert Mueller

    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
    302
    +0.11
    0.5%
    346
    +0.10
    0.4%
    538
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.11
    0.01
    200
    +0.10
    0.01
    302
    +0.10
    0.01
    Negative Logits
     Pyr
    -0.88
     Zoro
    -0.74
     Hawaiian
    -0.72
     Columb
    -0.71
    points
    -0.69
     Barbarian
    -0.69
     Syri
    -0.68
     Fiji
    -0.68
     Samurai
    -0.67
     Esper
    -0.67
    POSITIVE LOGITS
     Mueller
    0.96
     Probe
    0.89
     investigation
    0.87
    ueller
    0.86
    angelo
    0.86
     probe
    0.82
    owitz
    0.82
     investigations
    0.82
    croft
    0.79
     investigating
    0.78
    Activations Density 0.019%

    No Known Activations