© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 20548
    Prev
    Next
    INDEX
    Explanations

    terms related to solar power generation, including measurements like megawatts and solar energy projects

    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
    481
    +0.17
    0.8%
    447
    +0.13
    0.6%
    288
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.04
    537
    +0.13
    0.04
    678
    +0.11
    0.03
    Negative Logits
    ĸļ
    -0.84
    ozo
    -0.79
    bart
    -0.75
    nee
    -0.75
     Spiel
    -0.74
     McCann
    -0.72
    leans
    -0.72
    cus
    -0.71
    ãĥĥãĥī
    -0.70
    mone
    -0.70
    POSITIVE LOGITS
     installations
    1.15
    electric
    1.06
     plants
    1.03
     capacity
    1.01
     turbines
    1.00
     plant
    1.00
     electricity
    0.98
     infrastructure
    0.97
     towers
    0.96
     electr
    0.95
    Activations Density 0.105%

    No Known Activations