© 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. 13802
    Prev
    Next
    INDEX
    Explanations

    specific words related to information technology issues and controversies

    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
    447
    +0.21
    0.9%
    288
    +0.14
    0.6%
    314
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.21
    0.07
    575
    +0.14
    0.06
    651
    +0.10
    0.05
    Negative Logits
    ãĥĥãĥī
    -0.78
    ernels
    -0.71
    inas
    -0.71
    ooks
    -0.71
    tarians
    -0.70
    apps
    -0.67
     sexes
    -0.67
     discretion
    -0.66
    olars
    -0.66
     Steps
    -0.66
    POSITIVE LOGITS
     involving
    1.09
     called
    0.91
     titled
    0.90
     relating
    0.85
     wherein
    0.84
     entitled
    0.82
     whereby
    0.82
     aimed
    0.82
     consisting
    0.78
     dubbed
    0.78
    Activations Density 0.353%

    No Known Activations