© 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 49152
    9. 8-RES_FS49152-JB
    11. 45648
    Prev
    Next
    INDEX
    Explanations

    instances of the word "that"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.18
    0.8%
    679
    +0.16
    0.8%
    655
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    655
    +0.18
    0.06
    757
    +0.16
    0.05
    109
    +0.12
    0.05
    Negative Logits
     Doctrine
    -0.64
     Waiting
    -0.63
    Ire
    -0.62
    æĿ
    -0.62
    Knowing
    -0.61
     Situation
    -0.61
     Watching
    -0.61
    ONE
    -0.61
    channelAvailability
    -0.60
     VIDEOS
    -0.60
    POSITIVE LOGITS
     combines
    1.22
     incorporates
    1.10
     utilizes
    1.10
     specializes
    1.09
     originated
    1.09
     consists
    1.04
     extends
    1.04
     aims
    1.02
     operates
    1.01
     emphasizes
    1.01
    Activations Density 0.129%

    No Known Activations