© 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. 16405
    Prev
    Next
    INDEX
    Explanations

    discussions related to music and lyrics

    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
    456
    +0.13
    0.6%
    734
    +0.10
    0.4%
    23
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.04
    456
    +0.10
    0.02
    314
    +0.10
    0.00
    Negative Logits
     downstream
    -0.73
    BuyableInstoreAndOnline
    -0.70
     leash
    -0.70
     Cerberus
    -0.70
     respectively
    -0.69
     dumping
    -0.69
     worrying
    -0.66
     unrestricted
    -0.64
     consolidation
    -0.63
     butterflies
    -0.62
    POSITIVE LOGITS
    %:
    0.94
    maxwell
    0.91
    ENN
    0.91
    HAEL
    0.89
     laughs
    0.88
    Answer
    0.87
    laugh
    0.86
    miah
    0.82
    Interview
    0.82
    laughter
    0.80
    Activations Density 0.074%

    No Known Activations