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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 26789
    Prev
    Next
    INDEX
    Explanations

    references to the name "Nicholson" and variations, as well as the term "nob."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    421
    +0.13
    0.6%
    300
    +0.12
    0.6%
    52
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.01
    52
    +0.12
    0.01
    421
    +0.11
    0.01
    Negative Logits
     Luxem
    -0.82
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.78
     Shinra
    -0.77
     ultras
    -0.75
     mileage
    -0.72
    ãĥīãĥ©ãĤ´ãĥ³
    -0.70
     Miko
    -0.69
    RAY
    -0.68
    FACE
    -0.68
    DragonMagazine
    -0.68
    POSITIVE LOGITS
    ols
    1.24
    olas
    1.16
    olson
    1.14
    keye
    1.10
    olics
    1.08
    olls
    1.04
    ahon
    0.98
    ol
    0.97
    olic
    0.94
    ole
    0.92
    Activations Density 0.002%

    No Known Activations