© 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. 24100
    Prev
    Next
    INDEX
    Explanations

    specific terms and concepts related to coding and gaming, particularly in relation to Pokémon and other games

    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
    480
    +0.42
    2.1%
    447
    +0.12
    0.6%
    430
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.42
    -0.00
    266
    +0.12
    0.07
    326
    +0.11
    0.07
    Negative Logits
     Entered
    -0.80
     seism
    -0.54
     Gutenberg
    -0.53
     Pag
    -0.52
     asbestos
    -0.51
     Hutchinson
    -0.49
     Vaughan
    -0.49
     Townsend
    -0.49
     Windsor
    -0.48
     safegu
    -0.48
    POSITIVE LOGITS
     respectively
    0.66
    rect
    0.66
     itself
    0.66
    orea
    0.63
    ,,,,
    0.61
     alike
    0.59
    phia
    0.59
    eton
    0.57
    cycles
    0.57
    +.
    0.56
    Activations Density 0.766%

    No Known Activations