© 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 EvalsExportsinterp-engineNEWAPI 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. 4722
    Prev
    Next
    INDEX
    Explanations

    the word "are."

    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
    306
    +0.12
    0.6%
    169
    +0.11
    0.5%
    592
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.12
    0.05
    55
    +0.11
    0.04
    59
    +0.09
    0.03
    Negative Logits
    dom
    -0.93
    icism
    -0.87
    oire
    -0.83
    rouse
    -0.82
    odder
    -0.81
    matter
    -0.74
    uncture
    -0.74
    atform
    -0.72
    cture
    -0.70
    ashi
    -0.70
    POSITIVE LOGITS
    wolves
    1.10
    wolf
    0.93
    senal
    0.92
    types
    0.81
     idiots
    0.73
     miracles
    0.73
     examples
    0.72
     embodiments
    0.72
     illusions
    0.70
     jokes
    0.69
    Activations Density 0.132%

    No Known Activations