© 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. 5194
    Prev
    Next
    INDEX
    Explanations

    confessions and instances of confessing

    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
    571
    +0.11
    0.5%
    516
    +0.11
    0.5%
    343
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.11
    0.01
    288
    +0.11
    0.01
    343
    +0.10
    0.01
    Negative Logits
    chair
    -0.66
     braces
    -0.65
     stewards
    -0.64
     ripple
    -0.63
    chairs
    -0.62
     brackets
    -0.62
    axy
    -0.62
    gaard
    -0.60
    fare
    -0.60
     queues
    -0.60
    POSITIVE LOGITS
     confess
    1.11
     confessions
    1.07
     confession
    0.96
    itives
    0.91
     incrim
    0.91
     confessed
    0.89
    orial
    0.80
    essional
    0.71
     anonymously
    0.70
     decomp
    0.70
    Activations Density 0.026%

    No Known Activations