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. 679
    Prev
    Next
    INDEX
    Explanations

    medical precautions or conditions

    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
    367
    +0.13
    0.6%
    623
    +0.12
    0.5%
    717
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    717
    +0.13
    0.04
    367
    +0.12
    0.05
    623
    +0.10
    0.03
    Negative Logits
    ichen
    -0.76
    edin
    -0.70
    doms
    -0.69
    abases
    -0.68
    istries
    -0.66
    worth
    -0.65
    xus
    -0.63
    uez
    -0.63
    sites
    -0.62
    amy
    -0.62
    POSITIVE LOGITS
     progresses
    0.96
     accompan
    0.94
     basis
    0.89
     conduit
    0.85
     replacement
    0.85
     centerpiece
    0.85
     substitute
    0.81
     filler
    0.80
     precaution
    0.80
     identifier
    0.79
    Activations Density 0.280%

    No Known Activations