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 49152
    5. 8-RES_FS49152-JB
    6. 46055
    Prev
    Next
    INDEX
    Explanations

    the concept of functionality and successful operation in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    46
    +0.14
    0.7%
    423
    +0.11
    0.5%
    665
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.14
    0.06
    351
    +0.11
    0.03
    46
    +0.11
    0.04
    Negative Logits
    76561
    -0.87
     href
    -0.70
    é¾įå
    -0.69
    çīĪ
    -0.64
    cember
    -0.64
     deleg
    -0.64
     Barron
    -0.63
     denote
    -0.61
     vows
    -0.59
     railing
    -0.59
    POSITIVE LOGITS
     smoothly
    1.28
     efficiently
    1.14
     properly
    1.11
     reliably
    0.96
     uninterrupted
    0.96
     unim
    0.93
     correctly
    0.91
     seamlessly
    0.91
     faster
    0.90
     safely
    0.88
    Activations Density 0.144%

    No Known Activations