© 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. 4182
    Prev
    Next
    INDEX
    Explanations

    terms related to policies and their impact, particularly focusing on taxation and social programs

    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
    714
    +0.13
    0.6%
    600
    +0.10
    0.4%
    634
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.06
    123
    +0.10
    0.05
    529
    +0.10
    0.04
    Negative Logits
    DragonMagazine
    -0.71
     Mast
    -0.71
     hier
    -0.70
    canon
    -0.68
     Tag
    -0.66
     Fou
    -0.66
    cknowled
    -0.65
     vers
    -0.63
     subscript
    -0.62
     Parish
    -0.62
    POSITIVE LOGITS
     disproportionately
    1.04
     disadvantaged
    0.96
     beneficiaries
    0.95
    economic
    0.92
     taxpayers
    0.87
     wealthier
    0.85
    acea
    0.84
     economically
    0.84
     harms
    0.84
    elfare
    0.84
    Activations Density 0.421%

    No Known Activations