© 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. 9445
    Prev
    Next
    INDEX
    Explanations

    words related to political and economic issues, especially in the context of government regulation and policy

    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
    393
    +0.12
    0.5%
    274
    +0.10
    0.5%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.12
    0.04
    655
    +0.10
    0.04
    662
    +0.10
    0.04
    Negative Logits
    izont
    -0.84
    ãĤ´ãĥ³
    -0.76
    estern
    -0.67
    quer
    -0.67
    PRESS
    -0.66
    yna
    -0.64
    istor
    -0.63
    aukee
    -0.62
    ovember
    -0.62
    emis
    -0.61
    POSITIVE LOGITS
     consequently
    1.05
     they
    0.97
     furthermore
    0.93
     moreover
    0.89
    soever
    0.89
     therefore
    0.84
     includes
    0.81
     accordingly
    0.79
     reconc
    0.79
     there
    0.78
    Activations Density 0.176%

    No Known Activations