© 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. 5290
    Prev
    Next
    INDEX
    Explanations

    code snippets related to forum discussions on gaming characters and abilities

    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
    654
    +0.11
    0.5%
    272
    +0.11
    0.5%
    658
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    588
    +0.11
    0.04
    409
    +0.11
    0.03
    192
    +0.10
    0.03
    Negative Logits
     wast
    -0.83
     Yard
    -0.78
    ighed
    -0.76
     Beir
    -0.76
     Month
    -0.74
     Household
    -0.72
    igh
    -0.71
    ighth
    -0.70
    alls
    -0.70
    eah
    -0.69
    POSITIVE LOGITS
    ionics
    1.00
     Biology
    0.92
    etics
    0.91
    iology
    0.89
    ionic
    0.88
     biology
    0.86
     Programming
    0.83
     technology
    0.83
     Animation
    0.83
    Technology
    0.82
    Activations Density 0.333%

    No Known Activations