© 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. 19458
    Prev
    Next
    INDEX
    Explanations

    social media handles preceded by '@'

    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
    640
    +0.12
    0.5%
    669
    +0.10
    0.4%
    500
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    434
    +0.12
    0.03
    518
    +0.10
    0.03
    107
    +0.10
    0.02
    Negative Logits
     specificity
    -0.78
     vessels
    -0.75
     outnumbered
    -0.70
     lungs
    -0.69
     favor
    -0.68
     consolidation
    -0.68
     conver
    -0.68
     estab
    -0.68
     conformity
    -0.67
     favors
    -0.67
    POSITIVE LOGITS
    thereal
    1.13
    jon
    1.06
    Coach
    1.03
    jac
    1.01
    username
    0.98
    meg
    0.94
    sch
    0.93
    sam
    0.92
    jay
    0.91
    nat
    0.91
    Activations Density 0.023%

    No Known Activations