© 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. 9954
    Prev
    Next
    INDEX
    Explanations

    legal and bureaucratic processes

    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
    288
    +0.13
    0.6%
    644
    +0.12
    0.5%
    447
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    644
    +0.13
    0.03
    271
    +0.12
    0.03
    225
    +0.12
    0.03
    Negative Logits
    mson
    -0.65
    hiro
    -0.64
    azo
    -0.64
    birds
    -0.61
    arus
    -0.60
     Fighters
    -0.60
    aez
    -0.58
     spite
    -0.58
     hearty
    -0.58
    azines
    -0.58
    POSITIVE LOGITS
    ional
    1.28
    ions
    0.97
    ioned
    0.86
     whereby
    0.86
     process
    0.84
    ivism
    0.76
    ion
    0.76
    ogue
    0.72
    ESSION
    0.72
     outsourcing
    0.70
    Activations Density 0.028%

    No Known Activations