© 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. 15430
    Prev
    Next
    INDEX
    Explanations

    phrases related to emotional and impactful reactions, such as shock, fear, grief, and strong emotional responses in conversations

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    667
    +0.11
    0.5%
    481
    +0.10
    0.4%
    681
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    172
    +0.11
    0.05
    510
    +0.10
    0.04
    608
    +0.10
    0.05
    Negative Logits
    utical
    -0.75
    heses
    -0.72
    ourage
    -0.69
    uments
    -0.69
    ividual
    -0.68
     condu
    -0.68
    ãģ®éŃĶ
    -0.67
    gdala
    -0.67
    oln
    -0.67
    externalActionCode
    -0.65
    POSITIVE LOGITS
     hearing
    1.02
     witnessing
    1.01
     seeing
    0.97
     discovering
    0.94
     sudden
    0.88
     losing
    0.81
     prospect
    0.81
     sight
    0.80
     recalling
    0.80
     encountering
    0.79
    Activations Density 0.427%

    No Known Activations