© 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 EvalsExportsinterp-engineNEWAPI 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. 21137
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to politics and legal proceedings

    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
    481
    +0.10
    0.5%
    336
    +0.09
    0.4%
    15
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.10
    0.05
    15
    +0.09
    0.06
    683
    +0.09
    0.04
    Negative Logits
    mble
    -0.82
    lehem
    -0.75
    aughtered
    -0.72
    ories
    -0.68
    ãĥĩãĤ£
    -0.66
     Mehran
    -0.64
    ãĥIJ
    -0.64
    nz
    -0.62
    raints
    -0.62
    ãĤµ
    -0.61
    POSITIVE LOGITS
    's
    1.04
     himself
    0.93
     personally
    0.90
     candidacy
    0.81
     psychologically
    0.72
    ÃŃs
    0.69
     Himself
    0.68
    âĢ²
    0.68
     detractors
    0.67
     supporters
    0.65
    Activations Density 0.179%

    No Known Activations