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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 5905
    Prev
    Next
    INDEX
    Explanations

    topics related to discussing and addressing various societal issues like mental health, medical conditions, family dynamics, and social challenges such as obesity

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    481
    +0.12
    0.5%
    155
    +0.11
    0.5%
    393
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.12
    0.04
    723
    +0.11
    0.03
    648
    +0.10
    0.03
    Negative Logits
    OGR
    -0.75
    kie
    -0.73
    omers
    -0.71
    SPONSORED
    -0.71
    irl
    -0.70
    kai
    -0.70
    hiba
    -0.68
    Written
    -0.67
    lig
    -0.67
    ï¸ı
    -0.66
    POSITIVE LOGITS
     how
    1.02
     reforming
    0.99
     improving
    0.87
     topics
    0.87
     specifics
    0.86
     wanting
    0.82
     quitting
    0.82
     whether
    0.81
     preserving
    0.80
     overcoming
    0.79
    Activations Density 0.094%

    No Known Activations