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 24576
    5. 8-RES_FS24576-JB
    6. 23539
    Prev
    Next
    INDEX
    Explanations

    narratives related to personal struggles and health challenges

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    447
    +0.15
    0.7%
    486
    +0.11
    0.5%
    254
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    668
    +0.15
    0.04
    545
    +0.11
    0.04
    610
    +0.09
    0.03
    Negative Logits
     Governors
    -0.73
    ".[
    -0.73
     wink
    -0.67
    canon
    -0.67
    relevant
    -0.67
     implying
    -0.65
    notations
    -0.63
    ."[
    -0.63
     generals
    -0.62
    ).[
    -0.61
    POSITIVE LOGITS
     enrolled
    0.91
     enroll
    0.83
     biking
    0.81
     volunteering
    0.79
     bicy
    0.79
    guyen
    0.78
     hiking
    0.78
     surfing
    0.78
     Tinder
    0.77
     herself
    0.75
    Activations Density 0.623%

    No Known Activations