© 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. 8129
    Prev
    Next
    INDEX
    Explanations

    instances where the text instructs or addresses the reader to take specific actions or provides information about subscribing to newsletters

    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
    679
    +0.14
    0.6%
    707
    +0.13
    0.6%
    271
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    194
    +0.14
    0.02
    707
    +0.13
    0.02
    508
    +0.10
    0.02
    Negative Logits
    come
    -0.74
    jury
    -0.74
    bra
    -0.74
    had
    -0.69
    cu
    -0.68
     Strauss
    -0.68
    hua
    -0.67
    wikipedia
    -0.67
    acted
    -0.65
    phal
    -0.63
    POSITIVE LOGITS
     inbox
    1.03
     subscriptions
    1.01
     newsletter
    0.97
     newsletters
    0.96
     subscription
    0.95
     Subscribe
    0.94
     Seym
    0.92
     subscribers
    0.92
     Newsletter
    0.88
    Subscribe
    0.84
    Activations Density 0.017%

    No Known Activations