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 3072
    5. 8-RES_FS3072-JB
    6. 2692
    Prev
    Next
    INDEX
    Explanations

    sentences expressing personal emotions and experiences

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    362
    +0.21
    1.1%
    393
    +0.21
    1.0%
    481
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.21
    0.29
    393
    +0.21
    0.21
    608
    +0.14
    0.23
    Negative Logits
     thereby
    -0.86
     according
    -0.77
     aimed
    -0.76
    Allows
    -0.76
    LEASE
    -0.73
     namely
    -0.72
    thood
    -0.71
    perse
    -0.71
    require
    -0.69
    nesty
    -0.69
    POSITIVE LOGITS
     slightest
    1.27
     rest
    1.24
     whole
    1.19
     guy
    1.18
     entirety
    1.17
     coolest
    1.17
     proverbial
    1.11
     entire
    1.09
     same
    1.08
     smallest
    1.07
    Activations Density 2.424%

    No Known Activations