© 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. 7500
    Prev
    Next
    INDEX
    Explanations

    mentions of political and legal matters, including discussions around legislation, trials, and governmental responsibilities

    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
    373
    +0.30
    1.5%
    470
    +0.12
    0.6%
    608
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.30
    0.09
    680
    +0.12
    0.04
    470
    +0.11
    0.04
    Negative Logits
     goddamn
    -0.94
     darn
    -0.90
    .</
    -0.87
     ().
    -0.87
    .�
    -0.85
     someday
    -0.82
     dudes
    -0.77
    Pretty
    -0.77
     dude
    -0.76
     thereof
    -0.76
    POSITIVE LOGITS
     Nato
    1.05
     Labour
    1.04
     parliamentary
    1.03
     Fifa
    1.02
     manoeuv
    1.01
     criticised
    0.98
     Britons
    0.96
     campaigners
    0.95
     organisers
    0.94
     programme
    0.94
    Activations Density 1.409%

    No Known Activations