© 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. 9095
    Prev
    Next
    INDEX
    Explanations

    sentences referring to quantities or comparisons of multiple items

    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
    +0.14
    0.6%
    459
    +0.11
    0.5%
    613
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    395
    +0.14
    0.04
    71
    +0.11
    0.03
    344
    +0.10
    0.04
    Negative Logits
    assisted
    -0.77
    ãĥ¼ãĥ«
    -0.76
    buster
    -0.76
    anism
    -0.76
    livion
    -0.74
    venture
    -0.73
    gypt
    -0.72
    ADVERTISEMENT
    -0.71
    athon
    -0.70
    arenthood
    -0.70
    POSITIVE LOGITS
     reasons
    1.27
     similarities
    1.27
     ways
    1.25
     indications
    1.17
     occasions
    1.12
     exceptions
    1.10
     differences
    1.04
     examples
    1.04
     advantages
    1.03
     possibilities
    1.03
    Activations Density 0.101%

    No Known Activations