© 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. 12037
    Prev
    Next
    INDEX
    Explanations

    names of places or establishments, particularly related to country music or specific cities

    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
    288
    +0.12
    0.6%
    635
    +0.12
    0.5%
    737
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    737
    +0.12
    0.03
    97
    +0.12
    0.02
    718
    +0.12
    0.03
    Negative Logits
    inventoryQuantity
    -0.86
    acity
    -0.78
    unci
    -0.77
    ector
    -0.72
    ovember
    -0.71
    sbm
    -0.68
    wcsstore
    -0.67
    Accessory
    -0.67
    acious
    -0.67
    Vol
    -0.66
    POSITIVE LOGITS
    wide
    1.25
    oslov
    1.05
    side
    0.98
    men
    0.96
     Airlines
    0.83
     Laos
    0.75
    man
    0.73
     skiing
    0.72
     nationals
    0.70
    ESE
    0.69
    Activations Density 0.063%

    No Known Activations