© 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. 15042
    Prev
    Next
    INDEX
    Explanations

    descriptions of social issues related to urban areas or communities

    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
    109
    +0.10
    0.5%
    756
    +0.10
    0.5%
    372
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.10
    -0.01
    283
    +0.10
    0.03
    686
    +0.09
    0.05
    Negative Logits
    canon
    -0.83
     (@
    -0.79
     Osw
    -0.69
     tonight
    -0.67
    DK
    -0.67
     Verse
    -0.65
     Editor
    -0.64
     Yesterday
    -0.64
     Debate
    -0.64
    endum
    -0.62
    POSITIVE LOGITS
     cheaply
    1.06
     lucrative
    1.06
     themselves
    0.99
    their
    0.97
     their
    0.91
     coveted
    0.86
     inflated
    0.85
     costly
    0.84
     wealthier
    0.84
    entimes
    0.83
    Activations Density 1.053%

    No Known Activations