© 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. 5516
    Prev
    Next
    INDEX
    Explanations

    phrases related to observation or experience

    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
    447
    +0.11
    0.5%
    361
    +0.10
    0.5%
    481
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    349
    +0.11
    0.02
    403
    +0.10
    0.02
    240
    +0.10
    0.02
    Negative Logits
    quote
    -0.85
    starter
    -0.82
    ulum
    -0.73
    hp
    -0.72
    govtrack
    -0.70
    bi
    -0.69
    plan
    -0.69
    bags
    -0.67
    enthal
    -0.66
    oiler
    -0.66
    POSITIVE LOGITS
     firsthand
    1.42
     events
    0.87
    ewitness
    0.86
    Events
    0.85
     unfold
    0.84
     pandemonium
    0.83
     witnessing
    0.81
     demonstrations
    0.80
     unfolding
    0.77
     developments
    0.76
    Activations Density 0.056%

    No Known Activations