© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 10412
    Prev
    Next
    INDEX
    Explanations

    proper names of people and dates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    113
    +0.13
    0.6%
    530
    +0.11
    0.5%
    39
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.06
    530
    +0.11
    0.04
    39
    +0.11
    0.04
    Negative Logits
    soDeliveryDate
    -0.77
    bledon
    -0.71
     obstruct
    -0.67
     retake
    -0.63
     obstruction
    -0.63
     receipts
    -0.63
     worrying
    -0.62
     prey
    -0.61
     decre
    -0.61
     stalled
    -0.60
    POSITIVE LOGITS
     Contribut
    0.92
     Originally
    0.89
     (@
    0.87
     Edited
    0.87
     ·
    0.83
     Published
    0.83
     »
    0.81
    cerpt
    0.80
     Posted
    0.78
     ):
    0.78
    Activations Density 0.388%

    No Known Activations