© 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. 16203
    Prev
    Next
    INDEX
    Explanations

    personal pronouns and verbs related to personal experiences

    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
    +0.35
    1.8%
    481
    +0.13
    0.7%
    566
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.35
    0.10
    566
    +0.13
    0.09
    762
    +0.11
    0.09
    Negative Logits
     Millennium
    -0.65
     Innocent
    -0.59
     Skydragon
    -0.57
     Mandatory
    -0.56
    lihood
    -0.54
     Rothschild
    -0.52
    assembly
    -0.52
     Gene
    -0.51
     Clover
    -0.51
     Omega
    -0.50
    POSITIVE LOGITS
    've
    0.86
     reluct
    0.85
     struggled
    0.83
    'll
    0.79
     underwent
    0.79
     transitioned
    0.79
     encount
    0.79
     racked
    0.78
     embarked
    0.77
     experimented
    0.77
    Activations Density 0.601%

    No Known Activations