© 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. 2230
    Prev
    Next
    INDEX
    Explanations

    details related to regulations and legislation, particularly concerning exemptions and requirements

    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
    567
    +0.12
    0.6%
    575
    +0.10
    0.5%
    508
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.12
    0.04
    756
    +0.10
    -0.02
    570
    +0.10
    0.01
    Negative Logits
    wow
    -0.67
    !]
    -0.67
    )!
    -0.66
     snowball
    -0.63
     sadly
    -0.63
     Zeit
    -0.62
     disappoint
    -0.62
    Lind
    -0.61
     fortunately
    -0.61
     itch
    -0.60
    POSITIVE LOGITS
     exempted
    0.86
     ineligible
    0.85
     allowable
    0.84
     designated
    0.84
     complying
    0.84
     whichever
    0.83
     concess
    0.82
     compulsory
    0.82
     annually
    0.81
     permissible
    0.77
    Activations Density 0.751%

    No Known Activations