© 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. 12179
    Prev
    Next
    INDEX
    Explanations

    phrases involving past events or actions

    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
    +0.21
    1.0%
    458
    +0.11
    0.5%
    328
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.21
    0.03
    612
    +0.11
    0.02
    569
    +0.10
    0.02
    Negative Logits
    ĺħ
    -0.89
     nonetheless
    -0.75
    cellaneous
    -0.73
    âĹ¼
    -0.73
    respective
    -0.72
    ĵĺ
    -0.69
    verend
    -0.69
    cffff
    -0.67
     constantly
    -0.66
     nevertheless
    -0.65
    POSITIVE LOGITS
     Hoy
    0.63
     glance
    0.61
    ItemImage
    0.60
     sketches
    0.58
     suspicion
    0.57
    aeus
    0.57
    onso
    0.57
     Bean
    0.55
     whiff
    0.55
     Tanaka
    0.55
    Activations Density 0.206%

    No Known Activations