© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 15609
    Prev
    Next
    INDEX
    Explanations

    strong emotional expressions or reactions in the text

    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
    501
    +0.09
    0.4%
    81
    +0.09
    0.4%
    533
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.09
    0.06
    533
    +0.09
    0.04
    555
    +0.08
    0.04
    Negative Logits
     ][
    -0.68
    ortium
    -0.67
    esa
    -0.63
    ague
    -0.61
    ocument
    -0.60
    mite
    -0.60
    ibrary
    -0.60
    ocaust
    -0.60
    adelphia
    -0.59
    rez
    -0.59
    POSITIVE LOGITS
     accordingly
    1.26
     consequently
    0.88
     thereafter
    0.88
     alike
    0.86
     himself
    0.76
     afterward
    0.75
     therein
    0.74
     afterwards
    0.73
     consequ
    0.70
     therefore
    0.70
    Activations Density 0.354%

    No Known Activations