© 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. 8682
    Prev
    Next
    INDEX
    Explanations

    details of a crime scene and reactions or comments from witnesses or neighbors

    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
    +0.19
    0.9%
    87
    +0.12
    0.6%
    481
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.19
    0.05
    119
    +0.12
    0.04
    762
    +0.12
    0.06
    Negative Logits
    .</
    -0.70
    etheless
    -0.69
    .*
    -0.69
    ãĢĤ
    -0.68
    REDACTED
    -0.68
    ?).
    -0.67
    arist
    -0.66
    .).
    -0.64
    .(
    -0.62
    %.
    -0.62
    POSITIVE LOGITS
    ,"
    1.12
    ,'"
    1.06
     [
    1.04
    ,''
    0.98
    ),"
    0.89
    .,"
    0.85
    ,'
    0.79
    "—
    0.77
    ',"
    0.74
    "],
    0.73
    Activations Density 0.572%

    No Known Activations