© 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. 16951
    Prev
    Next
    INDEX
    Explanations

    references to constitutional rights, specifically the First Amendment

    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
    496
    +0.12
    0.5%
    394
    +0.11
    0.5%
    311
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    394
    +0.12
    0.02
    752
    +0.11
    0.02
    520
    +0.10
    0.01
    Negative Logits
     Io
    -0.83
    hin
    -0.75
    balls
    -0.74
    ynthesis
    -0.73
    cash
    -0.73
    DAQ
    -0.73
    DAY
    -0.73
    NYSE
    -0.71
    station
    -0.71
    hao
    -0.67
    POSITIVE LOGITS
     Rights
    1.05
     freedoms
    1.03
     rights
    1.00
     protections
    1.00
     Amendment
    0.96
     Amendments
    0.94
    endment
    0.90
     liberties
    0.86
     protection
    0.83
     protects
    0.82
    Activations Density 0.022%

    No Known Activations