© 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. 6061
    Prev
    Next
    INDEX
    Explanations

    discussions about planetary changes, particularly related to climate and environmental impact

    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
    480
    +0.14
    0.6%
    314
    +0.13
    0.6%
    298
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.14
    0.08
    298
    +0.13
    0.06
    351
    +0.09
    0.04
    Negative Logits
     reminds
    -0.80
    é¾į
    -0.60
     reminded
    -0.60
     Latest
    -0.60
     Favorite
    -0.59
     logos
    -0.56
     classy
    -0.56
    Tips
    -0.56
     sleek
    -0.55
     trademark
    -0.54
    POSITIVE LOGITS
     meaningful
    0.86
     satisfactory
    0.84
     satisf
    0.81
     desired
    0.81
    )).
    0.79
     materially
    0.79
    sufficient
    0.78
     worthwhile
    0.76
     ourselves
    0.76
     outcomes
    0.76
    Activations Density 0.996%

    No Known Activations