© 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. 1920
    Prev
    Next
    INDEX
    Explanations

    descriptions of technical specifications for electronic devices

    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
    406
    +0.11
    0.5%
    765
    +0.09
    0.4%
    343
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    765
    +0.11
    0.06
    406
    +0.09
    0.05
    49
    +0.09
    0.06
    Negative Logits
     Transcript
    -0.78
    azard
    -0.76
     contag
    -0.73
     sermon
    -0.70
     contagious
    -0.69
    chance
    -0.69
    pmwiki
    -0.68
     Soros
    -0.68
     Areas
    -0.64
    bart
    -0.63
    POSITIVE LOGITS
     equipped
    1.22
     chassis
    1.14
     fitted
    1.11
    ufact
    1.10
     prototypes
    1.08
     debuted
    1.08
     redesigned
    1.07
     specs
    1.05
     prototype
    1.05
     manufactured
    1.04
    Activations Density 0.955%

    No Known Activations