© 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 24576
    9. 8-RES_FS24576-JB
    11. 10054
    Prev
    Next
    INDEX
    Explanations

    the possessive form of names

    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
    578
    +0.13
    0.6%
    393
    +0.11
    0.5%
    575
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    578
    +0.13
    0.03
    664
    +0.11
    0.03
    28
    +0.10
    0.03
    Negative Logits
    GBT
    -0.93
    yz
    -0.80
    icity
    -0.77
    facts
    -0.73
    ĪĴ
    -0.71
    linked
    -0.70
    ļéĨĴ
    -0.70
    FY
    -0.69
    Ô
    -0.69
    ucl
    -0.69
    POSITIVE LOGITS
    haw
    0.94
     apprentice
    0.94
     remorse
    0.93
     dilemma
    0.91
     Wife
    0.89
     Guild
    0.89
     Guide
    0.88
     paradise
    0.86
     Creed
    0.85
     lounge
    0.83
    Activations Density 0.073%

    No Known Activations