© 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. 18384
    Prev
    Next
    INDEX
    Explanations

    words related to enhancement, improvement, and empowerment

    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
    184
    +0.11
    0.5%
    447
    +0.11
    0.5%
    665
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.05
    184
    +0.11
    0.03
    421
    +0.09
    0.04
    Negative Logits
    mad
    -0.87
    DragonMagazine
    -0.87
    thing
    -0.77
    sbm
    -0.74
    rants
    -0.69
    */(
    -0.67
    signed
    -0.66
    tics
    -0.65
    fuck
    -0.65
    nothing
    -0.64
    POSITIVE LOGITS
     productivity
    1.37
     efficiency
    1.37
     competitiveness
    1.23
     affordability
    1.15
     awareness
    1.09
     mobility
    1.06
     profitability
    1.06
     morale
    1.05
     longevity
    1.04
     Efficiency
    1.03
    Activations Density 0.145%

    No Known Activations