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. 40075
    Prev
    Next
    INDEX
    Explanations

    references to the word "virgin" and its variations

    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
    288
    +0.15
    0.7%
    92
    +0.11
    0.5%
    238
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.15
    0.01
    571
    +0.11
    0.01
    616
    +0.09
    0.01
    Negative Logits
     Grimoire
    -0.69
     middle
    -0.69
     Dying
    -0.65
    cedented
    -0.64
    uyomi
    -0.64
     Unch
    -0.64
     Solitaire
    -0.63
     Clover
    -0.63
     Pearce
    -0.62
     Schultz
    -0.62
    POSITIVE LOGITS
    ulent
    1.25
    gins
    1.22
    ulence
    1.17
    ality
    0.99
    iously
    0.99
    ted
    0.98
    anus
    0.97
    ally
    0.97
    idis
    0.95
    imon
    0.90
    Activations Density 0.006%

    No Known Activations