© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 12206
    Prev
    Next
    INDEX
    Explanations

    references to lunch-related activities and settings

    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
    459
    +0.11
    0.5%
    35
    +0.10
    0.5%
    129
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    11
    +0.11
    0.01
    129
    +0.10
    0.01
    469
    +0.09
    0.01
    Negative Logits
    ignty
    -0.78
    SPONSORED
    -0.67
    fitting
    -0.64
     CSI
    -0.62
     persecution
    -0.62
     Conquer
    -0.62
     infertility
    -0.61
     Painter
    -0.60
     empath
    -0.60
    isSpecialOrderable
    -0.59
    POSITIVE LOGITS
    time
    1.12
    bowl
    1.05
    aturday
    0.96
    eteria
    0.95
    wich
    0.94
    boxes
    0.92
    worms
    0.91
    bag
    0.91
    iday
    0.89
     buffet
    0.86
    Activations Density 0.014%

    No Known Activations