© 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. 1252
    Prev
    Next
    INDEX
    Explanations

    information related to incidents, accidents, and legal situations involving individuals

    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
    347
    +0.12
    0.5%
    258
    +0.10
    0.5%
    218
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.12
    0.07
    102
    +0.10
    0.07
    566
    +0.10
    0.06
    Negative Logits
    ivals
    -0.85
    phabet
    -0.80
     comparisons
    -0.75
    doms
    -0.72
    rities
    -0.71
     scenarios
    -0.69
     narratives
    -0.69
    ividual
    -0.69
    venge
    -0.68
     lifestyles
    -0.68
    POSITIVE LOGITS
     malfunction
    1.38
     jammed
    1.11
     worn
    1.11
     cracked
    1.04
    equipped
    1.01
     broken
    1.00
     rigged
    1.00
     confiscated
    0.99
     activated
    0.99
     repaired
    0.98
    Activations Density 0.342%

    No Known Activations