© 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. 10453
    Prev
    Next
    INDEX
    Explanations

    passages related to literary works or scripts

    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
    +0.18
    0.8%
    481
    +0.11
    0.5%
    458
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    698
    +0.18
    0.04
    540
    +0.11
    0.06
    326
    +0.09
    0.07
    Negative Logits
    quartered
    -0.75
    ikes
    -0.72
    atform
    -0.71
    iencies
    -0.70
    bred
    -0.69
    brand
    -0.69
     brokers
    -0.69
     trustees
    -0.68
    arers
    -0.68
     administrators
    -0.67
    POSITIVE LOGITS
     verse
    1.28
     lyrics
    1.28
     sung
    1.27
     poem
    1.22
     lyric
    1.20
     melody
    1.20
     songs
    1.14
     aloud
    1.13
     Songs
    1.12
     song
    1.10
    Activations Density 0.807%

    No Known Activations