© 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. 18610
    Prev
    Next
    INDEX
    Explanations

    words related to the condition of items/products, specifically focusing on the excellent or near-perfect condition of the item

    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
    288
    +0.14
    0.7%
    459
    +0.14
    0.6%
    496
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.01
    326
    +0.14
    0.01
    459
    +0.11
    0.01
    Negative Logits
     Ake
    -0.76
    Magikarp
    -0.65
    HUD
    -0.62
    ://
    -0.61
     Airbnb
    -0.60
     Methodist
    -0.58
     voc
    -0.57
     Incarn
    -0.56
     modem
    -0.56
     Dropbox
    -0.56
    POSITIVE LOGITS
     Seym
    1.49
    stones
    0.82
    ulously
    0.81
    rats
    0.81
    ér
    0.80
    eer
    0.80
    ril
    0.80
    marks
    0.79
    eers
    0.79
    mark
    0.78
    Activations Density 0.023%

    No Known Activations