© 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. 1736
    Prev
    Next
    INDEX
    Explanations

    phrases related to political candidacy and competition

    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
    430
    +0.11
    0.5%
    420
    +0.10
    0.5%
    469
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.11
    0.03
    422
    +0.10
    0.03
    725
    +0.10
    0.03
    Negative Logits
    »Ĵ
    -0.75
     Zhu
    -0.66
    ĪĴ
    -0.66
    catentry
    -0.66
    elf
    -0.65
    otaur
    -0.62
    EStream
    -0.58
    YR
    -0.58
    erity
    -0.57
    DragonMagazine
    -0.57
    POSITIVE LOGITS
     indefinitely
    0.93
    lishes
    0.84
    til
    0.76
     quo
    0.75
     till
    0.74
    until
    0.73
     limbo
    0.73
     forever
    0.71
     unchanged
    0.71
    oos
    0.71
    Activations Density 0.201%

    No Known Activations