© 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 EvalsExportsAPI 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. 10226
    Prev
    Next
    INDEX
    Explanations

    technological terms and concepts related to security and regulations

    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
    +0.13
    0.6%
    525
    +0.11
    0.5%
    479
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.13
    0.05
    288
    +0.11
    0.06
    325
    +0.11
    0.04
    Negative Logits
     brass
    -0.89
     toget
    -0.82
     elves
    -0.78
     showc
    -0.77
     legend
    -0.74
    natureconservancy
    -0.71
     queen
    -0.71
     tempt
    -0.70
     mog
    -0.70
     nostalgic
    -0.70
    POSITIVE LOGITS
     Syndrome
    1.35
     Protection
    1.25
     Reduction
    1.22
     Disorder
    1.20
     Ratio
    1.20
     Examination
    1.16
     Agreement
    1.16
     Assessment
    1.15
     Units
    1.15
     Relationship
    1.14
    Activations Density 0.246%

    No Known Activations