Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 315
    Prev
    Next
    INDEX
    Explanations

    punctuation marks (specifically commas)

    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
    398
    +0.09
    0.4%
    739
    +0.08
    0.4%
    100
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    398
    +0.09
    0.01
    418
    +0.08
    0.01
    739
    +0.08
    0.01
    Negative Logits
    kinson
    -0.71
    edom
    -0.70
    gow
    -0.68
    hest
    -0.68
     fronts
    -0.64
    abal
    -0.61
    estones
    -0.58
    ufact
    -0.57
    nesses
    -0.57
    herer
    -0.56
    POSITIVE LOGITS
     please
    0.81
    please
    0.77
    Sorry
    0.77
     Please
    0.75
    Please
    0.75
     nor
    0.72
    avascript
    0.67
    iframe
    0.64
     PLEASE
    0.64
    ãĤ®
    0.63
    Activations Density 0.023%

    No Known Activations