© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 9888
    Prev
    Next
    INDEX
    Explanations

    specific plans or strategies outlined in texts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    619
    +0.11
    0.5%
    509
    +0.11
    0.5%
    575
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.11
    0.04
    619
    +0.11
    0.04
    247
    +0.10
    0.04
    Negative Logits
     ILCS
    -0.63
    nces
    -0.62
    ADRA
    -0.62
     Stain
    -0.61
     charism
    -0.60
    socket
    -0.59
    athi
    -0.58
    poll
    -0.57
     Flavoring
    -0.57
    Cause
    -0.56
    POSITIVE LOGITS
    emaker
    1.03
    isphere
    0.98
    ters
    0.93
    escape
    0.87
    etary
    0.87
    ets
    0.80
    ologies
    0.80
     2030
    0.79
    ter
    0.78
    meal
    0.78
    Activations Density 0.084%

    No Known Activations