© 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 1536
    9. 8-RES_FS1536-JB
    11. 1021
    Prev
    Next
    INDEX
    Explanations

    profession-related words and titles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    +0.14
    0.6%
    91
    +0.12
    0.5%
    575
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.14
    0.13
    91
    +0.12
    0.12
    288
    +0.11
    0.13
    Negative Logits
    uld
    -0.74
    plet
    -0.64
     domains
    -0.62
     Soldiers
    -0.61
     VIDEOS
    -0.61
    ãĥ¼ãĥ³
    -0.60
    runners
    -0.59
    Xi
    -0.58
     increments
    -0.57
    edIn
    -0.57
    POSITIVE LOGITS
     extraord
    1.48
     specializing
    1.25
     who
    1.09
    ess
    1.09
     whose
    1.03
    digy
    0.91
    esses
    0.91
     laureate
    0.90
     whom
    0.90
    who
    0.85
    Activations Density 1.814%

    No Known Activations