© 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. 9321
    Prev
    Next
    INDEX
    Explanations

    words related to political and social issues, especially those surrounding conflicts and policies

    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
    480
    +0.13
    0.6%
    4
    +0.09
    0.5%
    86
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    351
    +0.13
    0.04
    526
    +0.09
    -0.02
    762
    +0.09
    0.08
    Negative Logits
     NX
    -0.70
     Pair
    -0.67
     LinkedIn
    -0.65
     clarification
    -0.63
     directional
    -0.63
     Feedback
    -0.62
     Advice
    -0.62
     Finder
    -0.62
    Adds
    -0.60
     upbeat
    -0.60
    POSITIVE LOGITS
     trillions
    1.06
     impover
    0.93
     tragically
    0.92
     billions
    0.90
     senseless
    0.89
     countless
    0.89
    ocaust
    0.88
     barbaric
    0.87
     impoverished
    0.85
     incarcer
    0.84
    Activations Density 1.553%

    No Known Activations