© 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 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. 7745
    Prev
    Next
    INDEX
    Explanations

    phrases related to law enforcement activities and incidents

    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
    639
    +0.10
    0.5%
    152
    +0.10
    0.4%
    585
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.10
    0.03
    31
    +0.10
    0.02
    152
    +0.10
    0.03
    Negative Logits
    opian
    -0.91
    figure
    -0.80
    ãĥ¼ãĥĨ
    -0.74
    sed
    -0.74
    ularity
    -0.71
    term
    -0.71
    lav
    -0.70
    cially
    -0.69
    cot
    -0.68
    olas
    -0.68
    POSITIVE LOGITS
     SWAT
    0.99
     brass
    0.89
     paramed
    0.85
     EMS
    0.85
     Sergeant
    0.85
     Department
    0.83
     sergeant
    0.82
     superintendent
    0.82
     Inspector
    0.81
     Officer
    0.80
    Activations Density 0.063%

    No Known Activations