© 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 EvalsExportsAPI 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. 13077
    Prev
    Next
    INDEX
    Explanations

    references to specific entities or concepts

    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
    159
    +0.13
    0.6%
    393
    +0.11
    0.5%
    562
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.13
    0.08
    589
    +0.11
    0.05
    384
    +0.09
    0.04
    Negative Logits
    ãĤ¹ãĥĪ
    -0.82
    ãĥĨ
    -0.74
    masters
    -0.72
    Īè
    -0.70
    ãĥĺ
    -0.69
    aneers
    -0.69
    asures
    -0.68
    aus
    -0.67
    acers
    -0.67
    ivas
    -0.66
    POSITIVE LOGITS
     week
    0.97
     trope
    0.93
     article
    0.90
     guy
    0.90
     gist
    0.87
     particular
    0.87
     sucker
    0.87
     nifty
    0.86
     month
    0.84
     year
    0.83
    Activations Density 0.184%

    No Known Activations