© 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 12288
    9. 8-RES_FS12288-JB
    11. 1708
    Prev
    Next
    INDEX
    Explanations

    words related to various technical terms, such as resources, probes, industries, spills, babies, DoS attacks, solutions, contracts, structure, and costumes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    447
    +0.16
    0.7%
    481
    +0.14
    0.6%
    679
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.16
    0.09
    537
    +0.14
    0.07
    330
    +0.13
    0.07
    Negative Logits
    Honest
    -0.74
    Recomm
    -0.69
    Fla
    -0.67
    Finish
    -0.67
    stown
    -0.64
    FTWARE
    -0.63
    Awesome
    -0.61
    ãĥĥãĥī
    -0.61
    noon
    -0.60
    ãĥª
    -0.60
    POSITIVE LOGITS
     are
    1.25
     cannot
    1.20
     aren
    1.18
     tend
    1.09
     were
    1.07
     rarely
    1.07
     tended
    1.04
     differ
    1.03
     typically
    1.03
     may
    1.03
    Activations Density 0.634%

    No Known Activations