© 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 EvalsExportsinterp-engineNEWAPI 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. 24372
    Prev
    Next
    INDEX
    Explanations

    terms related to gastronomy and culinary experiences

    oai_token-act-pair · gpt-4o-miniTriggered 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
    373
    +0.13
    0.6%
    199
    +0.11
    0.5%
    204
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.13
    -0.01
    738
    +0.11
    0.03
    701
    +0.10
    0.03
    Negative Logits
     Corker
    -0.70
    selection
    -0.65
     Gos
    -0.62
     Chattanooga
    -0.61
     Ort
    -0.60
     Hag
    -0.60
     Kinnikuman
    -0.59
     bye
    -0.59
     Kansas
    -0.58
    keepers
    -0.58
    POSITIVE LOGITS
    andise
    1.03
    uries
    0.91
    ritic
    0.89
    eric
    0.87
    ogical
    0.83
    chnology
    0.80
    olesterol
    0.79
    rees
    0.79
    INAL
    0.78
    ophen
    0.76
    Activations Density 0.082%

    No Known Activations