© 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. 18741
    Prev
    Next
    INDEX
    Explanations

    terms related to negative 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
    689
    +0.09
    0.4%
    98
    +0.09
    0.4%
    156
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    728
    +0.09
    0.03
    666
    +0.09
    0.04
    172
    +0.09
    0.04
    Negative Logits
    ellen
    -0.64
    eatures
    -0.63
    Different
    -0.61
    ptin
    -0.60
    cription
    -0.60
     cylinders
    -0.60
     adjusting
    -0.59
    meric
    -0.59
    ynthesis
    -0.57
    ready
    -0.57
    POSITIVE LOGITS
     embarrassment
    1.15
     endanger
    1.08
     harm
    1.04
    angering
    1.04
     jeopard
    1.03
     inconvenience
    0.97
     fate
    0.96
     havoc
    0.95
     tragedies
    0.95
     consequences
    0.93
    Activations Density 0.538%

    No Known Activations