© 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. 1394
    Prev
    Next
    INDEX
    Explanations

    software or technology-related updates and mentions of various technical tools or systems

    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
    480
    +0.28
    1.4%
    447
    +0.25
    1.2%
    481
    +0.18
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.28
    0.06
    480
    +0.25
    0.02
    464
    +0.18
    0.06
    Negative Logits
    ãģĦ
    -0.88
    âĹ¼
    -0.87
    é¾įå
    -0.79
    ãģį
    -0.75
    æ©
    -0.74
    ãĤĮ
    -0.72
    ãĤ´ãĥ³
    -0.71
    Ire
    -0.69
    ãģĹ
    -0.68
    ãģĤ
    -0.68
    POSITIVE LOGITS
    ¶
    0.70
    naire
    0.70
     âĵĺ
    0.69
    ↵↵
    0.67
     overview
    0.62
    Joined
    0.61
     heats
    0.60
     greets
    0.59
     guide
    0.58
     Allows
    0.57
    Activations Density 0.352%

    No Known Activations