© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 14110
    Prev
    Next
    INDEX
    Explanations

    specific locations and landmarks

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    325
    +0.14
    0.6%
    447
    +0.13
    0.6%
    510
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    667
    +0.14
    0.04
    325
    +0.13
    0.04
    232
    +0.11
    0.04
    Negative Logits
    authent
    -0.85
    advertisement
    -0.83
    pract
    -0.82
    requ
    -0.81
     compan
    -0.81
    diagn
    -0.81
    accompan
    -0.79
    rep
    -0.79
    compan
    -0.78
    exec
    -0.77
    POSITIVE LOGITS
     Island
    1.35
     Springs
    1.34
     Boulevard
    1.31
     Valley
    1.29
     Avenue
    1.28
     Park
    1.24
     Heights
    1.24
     Gardens
    1.23
     Creek
    1.23
     Station
    1.22
    Activations Density 0.154%

    No Known Activations