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. 21842
    Prev
    Next
    INDEX
    Explanations

    phrases related to knowledge and comprehension

    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
    158
    +0.11
    0.5%
    724
    +0.10
    0.5%
    86
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    724
    +0.11
    0.04
    540
    +0.10
    0.04
    225
    +0.10
    0.04
    Negative Logits
    sylv
    -0.79
    ratulations
    -0.72
    enegger
    -0.69
    ccoli
    -0.69
    incinn
    -0.68
     Parenthood
    -0.66
    packed
    -0.64
    rha
    -0.64
     Klux
    -0.62
    nir
    -0.61
    POSITIVE LOGITS
     thereof
    1.07
     of
    1.02
     regarding
    0.78
     view
    0.76
    of
    0.74
    rait
    0.73
     biases
    0.72
     interpretation
    0.72
     Of
    0.71
     assumes
    0.70
    Activations Density 0.111%

    No Known Activations