© 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. 37570
    Prev
    Next
    INDEX
    Explanations

    phrases related to control and efficiency in various contexts, particularly regarding resources or processes

    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
    +0.41
    2.1%
    480
    +0.23
    1.2%
    481
    +0.19
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.41
    0.07
    480
    +0.23
    0.03
    464
    +0.19
    0.07
    Negative Logits
     Vaugh
    -0.77
     Niet
    -0.67
    enegger
    -0.65
    Ire
    -0.62
     Seym
    -0.61
    ãĥ¯ãĥ³
    -0.59
    arthed
    -0.58
     Moroc
    -0.58
     Ezek
    -0.57
    agra
    -0.57
    POSITIVE LOGITS
    ¶
    0.78
     interval
    0.59
    lycer
    0.58
     gamma
    0.57
     ):
    0.56
     decrease
    0.55
     moderate
    0.55
     abortions
    0.55
     decreased
    0.53
     transient
    0.53
    Activations Density 0.420%

    No Known Activations