© 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. 18721
    Prev
    Next
    INDEX
    Explanations

    words related to lack of constraint or restriction

    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
    535
    +0.13
    0.6%
    422
    +0.10
    0.5%
    120
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    535
    +0.13
    0.01
    422
    +0.10
    0.01
    677
    +0.10
    0.01
    Negative Logits
    ++++
    -0.75
    issance
    -0.73
    aurus
    -0.71
    ococ
    -0.71
    oleon
    -0.71
    IENCE
    -0.71
     Hilton
    -0.68
    ilee
    -0.68
    ior
    -0.67
    psey
    -0.67
    POSITIVE LOGITS
    leaf
    1.13
    weed
    0.96
    cloth
    0.93
    nesses
    0.91
    ness
    0.87
     coupling
    0.87
    luster
    0.83
    fitting
    0.82
     cannon
    0.82
    lings
    0.82
    Activations Density 0.007%

    No Known Activations