© 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. 16537
    Prev
    Next
    INDEX
    Explanations

    phrases related to law enforcement, crime, and justice

    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
    87
    +0.25
    1.2%
    447
    +0.14
    0.7%
    305
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    305
    +0.25
    0.07
    87
    +0.14
    -0.03
    115
    +0.12
    0.05
    Negative Logits
    whatever
    -0.65
    ionics
    -0.64
    tarians
    -0.64
    channelAvailability
    -0.61
    bleacher
    -0.60
    DragonMagazine
    -0.60
     isEnabled
    -0.60
    "))
    -0.60
    Lenin
    -0.59
    format
    -0.59
    POSITIVE LOGITS
     recently
    0.80
     unsuccessfully
    0.74
     fiasco
    0.68
     2009
    0.65
     publicized
    0.65
     debacle
    0.64
     unsuccessful
    0.63
     infamous
    0.63
     famously
    0.62
     botched
    0.60
    Activations Density 1.000%

    No Known Activations