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. 43155
    Prev
    Next
    INDEX
    Explanations

    references to specific objects or elements within a document

    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
    87
    +0.28
    1.5%
    393
    +0.18
    1.0%
    362
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.28
    0.14
    393
    +0.18
    0.10
    608
    +0.15
    0.10
    Negative Logits
    ESA
    -0.65
    nesty
    -0.63
    furt
    -0.63
     nevertheless
    -0.62
    âĢł
    -0.62
     amid
    -0.62
    ward
    -0.61
     âĢº
    -0.60
    require
    -0.60
     nonetheless
    -0.60
    POSITIVE LOGITS
     slightest
    1.18
     same
    1.12
     whole
    1.10
     coolest
    1.08
     guy
    1.03
     entire
    0.97
     smallest
    0.96
     entirety
    0.96
     rest
    0.93
     proverbial
    0.90
    Activations Density 0.587%

    No Known Activations