© 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. 20324
    Prev
    Next
    INDEX
    Explanations

    names of individuals and organizations

    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
    642
    +0.13
    0.6%
    447
    +0.13
    0.6%
    481
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.05
    326
    +0.13
    0.04
    659
    +0.12
    0.04
    Negative Logits
    ivity
    -0.89
    verted
    -0.74
    imation
    -0.69
    rend
    -0.68
    ĪĴ
    -0.66
    ivism
    -0.65
    acters
    -0.64
    istics
    -0.63
    imate
    -0.63
    ded
    -0.62
    POSITIVE LOGITS
    useum
    0.81
    unicip
    0.79
    astern
    0.77
    ONEY
    0.74
    umbai
    0.71
    hattan
    0.70
    HAEL
    0.68
    uppet
    0.68
    ikhail
    0.68
    ullah
    0.68
    Activations Density 0.136%

    No Known Activations