© 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. 10861
    Prev
    Next
    INDEX
    Explanations

    technical and scientific terms related to physics and technology

    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
    +0.31
    1.6%
    314
    +0.17
    0.8%
    652
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.31
    -0.02
    756
    +0.17
    -0.02
    368
    +0.12
    0.05
    Negative Logits
     cens
    -0.67
     districts
    -0.64
     oversight
    -0.60
     principals
    -0.56
     jurisdiction
    -0.56
     censorship
    -0.54
     criminal
    -0.54
     martial
    -0.52
     DISTRICT
    -0.51
     intervene
    -0.50
    POSITIVE LOGITS
    ufact
    0.79
    prototype
    0.60
     itself
    0.60
    ãĤ´ãĥ³
    0.59
    ãĤ¨ãĥ«
    0.57
    hander
    0.56
    ãģ®éŃĶ
    0.56
    icter
    0.55
    prints
    0.55
    Sov
    0.54
    Activations Density 0.931%

    No Known Activations