© 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. 15397
    Prev
    Next
    INDEX
    Explanations

    mentions of famous names or titles

    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
    276
    +0.14
    0.7%
    160
    +0.13
    0.6%
    488
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.14
    0.03
    276
    +0.13
    0.03
    107
    +0.10
    0.03
    Negative Logits
     mathemat
    -0.96
     horizont
    -0.88
     myster
    -0.87
     incorpor
    -0.86
     charism
    -0.86
     simultane
    -0.83
     pestic
    -0.82
     therap
    -0.82
    undai
    -0.81
     explan
    -0.79
    POSITIVE LOGITS
    ï¸ı
    1.09
    MQ
    1.00
    âĹ¼
    0.98
    ãĥĥãĥī
    0.97
    é¾į
    0.91
    deg
    0.90
    rd
    0.87
    horn
    0.83
    ove
    0.79
    Īè
    0.78
    Activations Density 0.031%

    No Known Activations