© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 18335
    Prev
    Next
    INDEX
    Explanations

    specific text patterns related to computer programming and technical instructions

    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
    314
    +0.14
    0.7%
    512
    +0.09
    0.4%
    519
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.14
    0.04
    739
    +0.09
    0.01
    6
    +0.09
    0.03
    Negative Logits
     warranties
    -0.62
     opinions
    -0.60
     invested
    -0.60
     Citiz
    -0.59
     wiser
    -0.58
     sentiments
    -0.57
    andise
    -0.57
     glean
    -0.57
     outraged
    -0.56
     ardent
    -0.56
    POSITIVE LOGITS
    20439
    0.82
    gged
    0.78
    士
    0.78
    rontal
    0.73
    ãĥ©ãĥ³
    0.72
    ãĥ¯ãĥ³
    0.72
    ħĭ
    0.71
    ãĥ´ãĤ¡
    0.70
    dated
    0.68
    ategory
    0.67
    Activations Density 0.239%

    No Known Activations