© 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. 1402
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to software applications and technology

    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
    447
    +0.19
    0.9%
    343
    +0.14
    0.7%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.19
    0.02
    592
    +0.14
    0.02
    288
    +0.10
    0.02
    Negative Logits
    omaly
    -0.71
    este
    -0.71
    rued
    -0.69
    iery
    -0.68
    itiz
    -0.67
    enegger
    -0.67
    geries
    -0.66
    jury
    -0.65
    metics
    -0.65
    notations
    -0.65
    POSITIVE LOGITS
    ®
    0.77
    Chat
    0.71
     Labs
    0.67
    Champ
    0.65
    loo
    0.63
    Connection
    0.63
    Hash
    0.62
     Kills
    0.61
     Messenger
    0.61
     Remastered
    0.60
    Activations Density 0.030%

    No Known Activations