© 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. 15549
    Prev
    Next
    INDEX
    Explanations

    phrases related to empathy or sympathy towards certain groups or situations

    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
    693
    +0.10
    0.5%
    561
    +0.10
    0.5%
    161
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.10
    0.01
    161
    +0.10
    0.01
    259
    +0.10
    0.01
    Negative Logits
    stall
    -0.77
    grad
    -0.72
    ouver
    -0.71
    corn
    -0.70
    teen
    -0.69
    growth
    -0.65
    clamation
    -0.65
    gradation
    -0.65
    paper
    -0.64
    fleet
    -0.63
    POSITIVE LOGITS
     sympath
    0.96
     sympathy
    0.87
     sympathetic
    0.81
    axter
    0.81
    athetic
    0.81
     Cosponsors
    0.81
    PsyNetMessage
    0.74
     empath
    0.74
     Networks
    0.74
     vein
    0.71
    Activations Density 0.025%

    No Known Activations