© 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. 167
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal actions and compliance issues

    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
    204
    +0.15
    0.7%
    245
    +0.11
    0.5%
    352
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.15
    0.06
    230
    +0.11
    0.04
    204
    +0.11
    0.04
    Negative Logits
     Appears
    -0.75
    okingly
    -0.72
    nis
    -0.69
    loo
    -0.67
    eday
    -0.65
    terson
    -0.64
    nen
    -0.62
    çīĪ
    -0.62
    Medium
    -0.61
    Chat
    -0.61
    POSITIVE LOGITS
     adequately
    1.16
     adequate
    1.01
     anymore
    0.95
     properly
    0.91
     nor
    0.84
     sufficiently
    0.83
     timely
    0.83
    due
    0.80
     sufficient
    0.79
     deadlines
    0.77
    Activations Density 0.344%

    No Known Activations