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

    Neuronpedia

    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. 19654
    Prev
    Next
    INDEX
    Explanations

    references to health care policy and insurance coverage, especially related to low-income Americans

    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
    447
    +0.13
    0.6%
    39
    +0.12
    0.6%
    481
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.13
    0.08
    51
    +0.12
    0.07
    762
    +0.11
    0.07
    Negative Logits
    yss
    -0.87
    Reloaded
    -0.79
    OLOG
    -0.74
    rawdownloadcloneembedreportprint
    -0.73
    saf
    -0.70
    mosp
    -0.69
    POR
    -0.66
     acknow
    -0.64
    iosyncr
    -0.64
    inventoryQuantity
    -0.64
    POSITIVE LOGITS
     alike
    1.24
    hips
    1.18
    hip
    1.04
     who
    1.03
    paces
    0.95
    pace
    0.94
     whom
    0.87
     aged
    0.86
    ervatives
    0.86
    opausal
    0.86
    Activations Density 0.291%

    No Known Activations