© 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. 2219
    Prev
    Next
    INDEX
    Explanations

    commands and instructions in technical documentation

    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
    745
    +0.15
    0.7%
    718
    +0.13
    0.6%
    373
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.15
    0.07
    49
    +0.13
    0.06
    508
    +0.12
    0.05
    Negative Logits
    ONSORED
    -0.73
     Attention
    -0.69
     Said
    -0.67
     Worth
    -0.63
     succumbed
    -0.63
     Salam
    -0.62
    IED
    -0.61
     costing
    -0.61
    living
    -0.60
     patri
    -0.60
    POSITIVE LOGITS
     customize
    1.39
    ggles
    1.38
     modify
    1.26
     configure
    1.26
     calculate
    1.25
     activate
    1.24
     simulate
    1.24
     manually
    1.22
     reproduce
    1.21
     obtain
    1.20
    Activations Density 0.160%

    No Known Activations