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. 569
    Prev
    Next
    INDEX
    Explanations

    adjectives reflecting a sense of strength or intensity

    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
    247
    +0.13
    0.6%
    101
    +0.11
    0.5%
    745
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.13
    0.04
    387
    +0.11
    0.05
    629
    +0.10
    0.04
    Negative Logits
     Sloan
    -0.71
    Hop
    -0.70
     Correction
    -0.69
     Newly
    -0.67
     McCann
    -0.66
     Hilton
    -0.66
     Wonderland
    -0.66
    artment
    -0.65
     Kare
    -0.65
    cision
    -0.63
    POSITIVE LOGITS
    nesses
    0.97
     enough
    0.94
    ener
    0.87
    enough
    0.85
     cryptography
    0.85
    ament
    0.84
     motiv
    0.80
     circumst
    0.77
     bonds
    0.76
     believer
    0.76
    Activations Density 0.723%

    No Known Activations