© 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. 15524
    Prev
    Next
    INDEX
    Explanations

    elements related to urban settings and activities, such as cars, streets, buildings, and events happening in the city

    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
    413
    +0.12
    0.5%
    114
    +0.11
    0.5%
    537
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    537
    +0.12
    0.07
    258
    +0.11
    0.07
    413
    +0.10
    0.06
    Negative Logits
    ãĤ´ãĥ³
    -0.72
    ãĥį
    -0.72
    âĸ¬
    -0.71
     experien
    -0.71
     Tycoon
    -0.68
     unification
    -0.67
     Identity
    -0.67
    history
    -0.66
     practition
    -0.66
    Reviewer
    -0.66
    POSITIVE LOGITS
    hips
    1.22
    creen
    1.10
    poons
    1.09
    leeve
    1.06
     parked
    1.03
    avers
    1.00
     lined
    1.00
    hops
    0.99
     dotted
    0.99
     overflowing
    0.99
    Activations Density 0.276%

    No Known Activations