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. 26103
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing the concept of 'the'

    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
    393
    +0.12
    0.6%
    572
    +0.11
    0.5%
    659
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.12
    0.03
    331
    +0.11
    0.02
    142
    +0.10
    0.02
    Negative Logits
    accompan
    -0.72
     includ
    -0.71
    isol
    -0.67
    respective
    -0.66
     deval
    -0.64
    netflix
    -0.64
    Gallery
    -0.63
    Pg
    -0.63
     amount
    -0.63
    steamapps
    -0.63
    POSITIVE LOGITS
     Georgetown
    0.93
     Syracuse
    0.93
     Bellev
    0.87
     Princeton
    0.84
     Harvard
    0.82
     Honolulu
    0.81
     Providence
    0.81
     Philadelphia
    0.79
     Seattle
    0.79
     Dallas
    0.78
    Activations Density 0.035%

    No Known Activations