© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 6699
    Prev
    Next
    INDEX
    Explanations

    phrases related to business tasks and advice

    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
    +0.15
    0.7%
    397
    +0.12
    0.5%
    633
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.06
    132
    +0.12
    0.04
    325
    +0.11
    0.04
    Negative Logits
    actionDate
    -0.74
    uay
    -0.71
    named
    -0.67
    ãĤ¨ãĥ«
    -0.66
     looph
    -0.66
    ortium
    -0.66
     Cosponsors
    -0.65
    rero
    -0.63
    farious
    -0.62
    cient
    -0.60
    POSITIVE LOGITS
     specials
    0.94
     festivities
    0.87
    stice
    0.78
    pour
    0.73
    idays
    0.73
    endars
    0.71
     curfew
    0.70
    wake
    0.68
     schedule
    0.68
     anniversary
    0.67
    Activations Density 0.438%

    No Known Activations