© 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. 4528
    Prev
    Next
    INDEX
    Explanations

    information related to medical research, scientific studies, and academia

    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
    620
    +0.11
    0.5%
    701
    +0.10
    0.5%
    765
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    425
    +0.11
    0.04
    765
    +0.10
    0.04
    620
    +0.10
    0.02
    Negative Logits
    fw
    -0.68
     fade
    -0.68
    ãĥ¥
    -0.68
    ize
    -0.65
    dule
    -0.65
     Pastebin
    -0.65
    ãĥ¼
    -0.65
     bleed
    -0.64
    sold
    -0.64
     heater
    -0.64
    POSITIVE LOGITS
     studying
    0.88
     specializing
    0.86
     Authors
    0.79
    igham
    0.79
     studied
    0.77
     professor
    0.77
     PhD
    0.76
    doctoral
    0.76
    inguished
    0.73
     adjunct
    0.72
    Activations Density 0.158%

    No Known Activations