Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 43443
    Prev
    Next
    INDEX
    Explanations

    verbs related to serving food or dishes

    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
    469
    +0.12
    0.5%
    174
    +0.10
    0.4%
    526
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.12
    0.01
    332
    +0.10
    0.01
    529
    +0.10
    0.01
    Negative Logits
     Nanto
    -0.82
    orph
    -0.72
    peria
    -0.71
     converter
    -0.70
    iqueness
    -0.69
    usting
    -0.69
    vu
    -0.65
    usted
    -0.63
    pora
    -0.63
    edia
    -0.62
    POSITIVE LOGITS
     meals
    0.91
     servings
    0.86
    cloth
    0.84
     Delicious
    0.81
     drinks
    0.79
    washer
    0.78
    iments
    0.77
    maid
    0.75
     breakfast
    0.74
     dinners
    0.74
    Activations Density 0.017%

    No Known Activations