© 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 12288
    9. 8-RES_FS12288-JB
    11. 10726
    Prev
    Next
    INDEX
    Explanations

    official titles and specific identifiers mentioned in news articles

    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
    447
    +0.14
    0.6%
    660
    +0.10
    0.5%
    92
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    660
    +0.14
    0.06
    224
    +0.10
    0.06
    91
    +0.10
    0.07
    Negative Logits
     Pieces
    -0.76
     instruments
    -0.75
    ories
    -0.75
    ovies
    -0.74
    ernels
    -0.73
    ooks
    -0.72
     colours
    -0.71
     calendars
    -0.70
    ippers
    -0.70
    rams
    -0.69
    POSITIVE LOGITS
     named
    1.23
     who
    1.14
     surn
    1.02
     testified
    1.00
     whose
    0.98
     nicknamed
    0.97
    who
    0.93
     whom
    0.93
     specializing
    0.87
    friend
    0.85
    Activations Density 0.354%

    No Known Activations